in Venture Bridging Theory and Practice

2020

Some of the Year’s Best Reads Year’s the Some of Industry Analysis Venture Digest that Deter Entrepreneurs? Entrepreneurs? Deter that Digital Arm the for in A Shot Health Innovation What are the Entrepreneurship Myths Entrepreneurship the are What Trends in Venture Trends The COVID-19 Crisis for VC – The COVID-19 VC for Crisis Death or Knell Springboard? Virtual Roundtable Is Artificial AI more Intelligent? than Deep Innovation Venture Policy and Management Policy Venture Economy? Venture the Killing IPAre Restrictions Coller School of Management School of Coller University Aviv Tel

Venture Review Coller

Coller Venture Review | 2020 Letter from the Editor

oller Venture Review, the flagship Over the last year, our Editorial Board has journal of the Coller Institute of become more involved and we are grateful that Venture at Tel Aviv University, they have contributed to our Venture Digest, C continues its mission to help highlighting some of the year’s best reads in areas bridge theory with practice in the areas of ranging from entrepreneurial team formation venture, innovation, and entrepreneurship. to social entrepreneurship. Many thanks to Prof. Shai Bernstein in particular for his leadership In this issue, we draw on the insights and of our virtual roundtable with VC investors. expertise of a range of contributors across Many thanks as well to Dr. Leslie Broudo, our changing technology paradigms and industries. Managing Editor. We appreciate the input of all Our articles continue to articulate emerging our contributors, colleagues, and collaborators trends, extract generalizable themes, and lend worldwide for their dedication and vision. insights associated with the codification of new ways of thinking linked to action in the We invite our colleagues to continue to conceptualization, financing, and execution follow us. As always, we remain focused on of innovation and new venture creation. our mission bridging theory and practice in venture in service of a shared and bright future As 2020 ends, I write with profound recognition ahead. While the results of our work will not of the challenges that the COVID-19 pandemic be measurable in weeks or months, we hope has brought. Consequently, the articles included this first step can help guide our future. We in this issue of the Coller Venture Review indeed welcome any comments and suggestions from reflect the global challenges and opportunities our readers that will help us improve the value across the changing venture landscape. We have of the Coller Venture Review to its readership. attempted to address the profound changes we have absorbed globally in every aspect of our lives, from digital health to fintech in a post- Sincerely, COVID-19 world. For entrepreneurs, investors, academic leaders, and CEOs the impact will transcend how we invest, what we invest in, how we work together, and the impact we can reasonably expect. We know that the choices will necessarily be practical as well as moral. Moshe Zviran Editor-in-Chief

01 Coller 4 36 70 Venture Policy and Management Virtual Roundtable Industry Analysis Venture Are IP Restrictions Killing the The COVID-19 Crisis for VC – A Shot in the Arm for Digital Review Venture Economy? Death Knell or Springboard? Health Innovation 7 IP, Power, and the Pandemic 39 How the COVID-19 Crisis Ignited 72 Leading Transformative Change in Joseph E. Stiglitz and Accelerated Venture Capital Digital Health – Lessons from Practice University Professor, Columbia University Fiona Darmon Lilach Weisz Nobel Memorial Prize Laureate in Economic Sciences General Partner, Head of Innovation and Tech Transfer, Jerusalem Venture Partners Tel-Aviv Sourasky Medical Center Arjun Jayadev Associate Professor of Economics, Felda Hardymon Tamar Many University of Massachusetts, Boston Partner, Bessemer Venture Partners, Co-founder, MindState Professor of Management Practice, Senior Design Lecturer, Achal Prabhala Harvard Business School Shenkar College of Engineering, Design and Art Fellow, Shuttleworth Foundation Shai Bernstein Associate Professor in Entrepreneurial Management, 13 The Curious Case of Patent Harvard Business School Balance Sheet Invisibility John A. Squires, Esq. 46 Partner, Dilworth Paxson, LLC; A New VC Fund Rides former Head of Intellectual Property, Goldman Sachs & Co. the Silver Tsunami Abby Levy David N. Lawrence, Esq. Managing Partner & Co-Founder, Founder, Risk Assistance Network + Exchange; PrimeTime Partners Former Managing Director, Business Intelligence, Goldman Sachs & Co. 18 56 78 Deep Innovation Trends in Venture Venture Digest Is AI more Artificial than Intelligent? What are the Entrepreneurship Myths Some of the Year’s Best Reads that Deter Entrepreneurs? 21 Artificial Intelligence: Sifting the Facts Aya Soffer 59 Shifting the Entrepreneurial Paradigm VP, AI Technology, IBM Global Research to a Data-Driven View 84 Kartik Hosanagar Ethan Mollick Professor of Marketing, Associate Professor of Management, John C. Hower Professor of Technology and Digital Business, Ralph J. Roberts Distinguished Faculty Scholar, Advisory Board Wharton School of the University of Pennsylvania Wharton School of the University of Pennsylvania

32 The Surprising Role that 67 FinTech and the Future of Banking AI Plays in Management Xavier Vives Professor of Economics and Finance, Lior Zalmanson IESE Business School Senior Lecturer of Technology and Information Management, Coller School of Management, Tel Aviv University

Gal Oestreicher-Singer Coller School of Management Professor of Technology and Information Management, Tel Aviv University Coller School of Management, Tel Aviv University

03 Overview

i ur Venture Policy and Management section 7 frames questions at the intersection of new venture creation and policy globally. In IP, Power, and the Pandemic Othis issue, we address some of the questions Professor Joseph E. Stiglitz surrounding intellectual property, including those brought University Professor, Columbia University about by COVID-19. Nobel Memorial Prize Laureate in Economic Sciences Joseph Stiglitz and his co-authors, Arjun Jayadev and Achal Prabhala, address the potential benefits of patent Dr. Arjun Jayadev pooling to support the more efficient development and Associate Professor of Economics, delivery of vaccines. University of Massachusetts, Boston Venture Policy and John A. Squires and David N. Lawrence, formerly of Mr. Achal Prabhala Goldman Sachs, address the non-systematic valuation Fellow, Shuttleworth Foundation of patents on the balance sheets of technology startups. Together, these articles combine theory and practice to help Management us consider how seeming individual-level changes become aggregated and amplified. They suggest both promise and 13 shifts in policy and regulation to ensure the distribution of The Curious Case of Patent benefits. From the perspective of both public health and global Are IP Restrictions Killing Balance Sheet Invisibility M&A, it appears clear that intellectual property measures of efficacy call for a systematic revaluation across current John A. Squires, Esq. geographic, demographic, and economic boundaries. Partner, Chairman of Emerging Company Practice, Looking forward, future discussions in the Venture Policy the Venture Economy? Dilworth, Paxson LLC and Management section will continue to raise important Former Head of Intellectual policy questions in keeping with trends in innovation and Property at Goldman Sachs & Co. new venture creation globally.

David N. Lawrence, Esq. Founder and Chief Collaborative Officer, Risk Assistance Network + Exchange (RANE) Former Associate General Counsel, Managing Director, and Head of Business Intelligence, Goldman Sachs & Co.

04 COLLER VENTURE REVIEW 05 IP, Power, and the Pandemic

Professor Joseph E. Stiglitz Mr. Achal Prabhala University Professor, Fellow, Shuttleworth Foundation Columbia University Nobel Memorial Prize Laureate in Economic Sciences

Dr. Arjun Jayadev Associate Professor of Economics, University of Massachusetts, Boston

magine a world in which a global discuss the latest data on emerging flu network of medical professionals strains, and to decide which strains monitored for emerging strains should be included in each year’s Iof a contagious virus, periodically vaccine. As a network of laboratories updated an established formula spanning 110 countries, funded almost for vaccinating against it, and then entirely by governments (and partly made that information available to by foundations), GISRS epitomizes companies and countries around the what Amy Kapczynski of Yale Law world. Moreover, imagine if this work School calls2 “open science.” were done without any intellectual- Because GISRS is focused solely property (IP) considerations, on protecting human lives, rather and without pharmaceutical than turning a profit, it is uniquely monopolies exploiting a desperate capable of gathering, interpreting, and public to maximize their profits. distributing actionable knowledge for Imagine too a world in which a the development of vaccines. While global network of scientists searched this approach may have been taken for vaccines and therapeutics to for granted in the past, its advantages combat COVID-19, with only an are quickly becoming clear. ambition of getting the medicines The world has changed a lot since Jonas to as many people as cheaply and Salk’s polio vaccine, which was made as quickly as possible—a world freely available immediately. Today in which the drug companies see most vaccines that come to market COVID-19 not as an opportunity are patented. For example, PCV13, for unprecedented profits, but as one the current multi-strain pneumonia for providing unprecedented benefits vaccine administered to babies, costs to a world immersed in a pandemic. hundreds of dollars because it is the This may sound like a utopian fantasy, monopoly property of Pfizer. And but it is actually a description of how although Gavi, the Vaccine Alliance the flu vaccine has been produced subsidizes some of the costs of the for the past 50 years. Through the vaccine in developing countries, World Health Organization’s Global many people still cannot afford it. In US$5 billion Influenza Surveillance and Response India, more than 100,000 preventable System1, experts from around the world infant deaths3 from pneumonia are Annual revenue for Pfizer from convene twice a year to analyze and recorded every year, while the • pneumonia vaccine

06 COLLER VENTURE REVIEW 07 vaccine brings in roughly $5 billion Patent pooling is not a new idea. in revenue4 for Pfizer annually. Through the Medicines Patent Pool5, the United Nations and the WHO As public-health advocates We need to acknowledge that the have for years sought to increase current system—in which private and scholars have long access to treatments for HIV/AIDS, monopolies profit from knowledge hepatitis C, and tuberculosis, and argued, monopolies kill, by that is largely produced by public have now expanded that program institutions—is not fit for purpose. As denying access to life-saving to cover COVID-19. Patent pools, public-health advocates and scholars prize funds, and other similar medicines that otherwise have long argued, monopolies kill, ideas are part of a broader agenda by denying access to life-saving would have been available to reform how life-saving drugs medicines that otherwise would have are developed and made available. under an alternative system been available under an alternative The goal is to replace a monopoly- system—like the one facilitating the driven system with one based on yearly production of the flu vaccine. cooperation and shared knowledge. their monopoly profits—leading to There is already some movement in In the current climate of cooperation less access and higher prices, and in favor of alternative approaches. For in the name of societal well-being, example, Costa Rica’s government some cases, to unnecessary deaths. it is easy to forget that in the recently called on the WHO to The insulin crisis in the US is only one pharmaceutical arena, what we have establish a voluntary pool of IP rights example of this. A drug that has been seen in the past is the very opposite for COVID-19 treatments, which well-known for decades is unaffordable of this. Pharmaceutical companies would allow multiple manufacturers for many primarily because of the have been involved in what can be to supply new drugs and diagnostics indiscriminate granting of follow called the “Enclosing of the Knowledge at more affordable prices. on patents. While there are instances Commons”, extending control over life- where drug companies have seen saving drugs through either frivolous remarkable advances, almost always or secondary patents, preventing their products rest on advances in production and use of generics, in basic science financed by governments, an attempt to indefinitely extend foundations, and education institutions. Most importantly, we don’t need to pay the enormous price that our broken system of intellectual property extracts to get these advances. Of course, some will try to argue that dealing with COVID-19 is a matter that is sui generis. They want to fall back just on pressuring the drug companies to behave well, not to charge excessively, with the threat of using compulsory licenses if they don’t. But Patent pools, prize funds, that would be a mistake: The COVID-19 crisis is simply exposing, in a dramatic and other similar ideas are way, the flaws in our current system. part of a broader agenda to At a time when cooperation is critical, we have an IP system that encourages reform how life-saving drugs secrecy; at a time when having the are developed and made widest dissemination of drugs at the lowest prices is essential for the public available. The goal is to replace health and societal well-being, we have a monopoly-driven system with an IP system that encourages charging what the market would bear—and one based on cooperation and right now, that’s an enormous amount. shared knowledge Fortunately, the scientific community has seen beyond short-term concerns about profits, but it is not clear that our pharmaceutical companies will, and even if they do for the moment, •

08 COLLER VENTURE REVIEW 09 by contrast has only been able to obtain a commitment of one dose per 9 people. Some vaccine producers have already For too long, we have bought into the announced that they will charge prices that are high enough to seriously myth that today’s IP regime is necessary. deter usage in poorer countries. Apart Academics and policymakers have from the obvious ethical concerns of this situation, the fact that doses already come forward with many About are not available for a contagious promising proposals for generating disease is likely to be deeply socially inefficient since it would prolong socially useful—rather than merely Professor Joseph E. Stiglitz is an award- winning American economist and a the pandemic with all its attendant profitable—pharmaceutical innovation University Professor at Columbia University. spillover externalities to production. He is the co-chair of the High-Level Expert The obvious solution here, too, Group on the Measurement of Economic is to promote what has come to be Performance and Social Progress at the known as a ‘people’s vaccine’8— OECD, and the Chief Economist of the the widespread and free usage of Roosevelt Institute. A recipient of the Nobel Memorial Prize in Economic vaccines, treatments and tests with Sciences and the John Bates Clark Medal, a priority given to the vulnerable, he is a former Senior Vice President frontline workers and poorer countries and Chief Economist of the World Bank with limited capacity to save lives. and a former member and chairman of But once again, that would require it is unlikely that they will do so over 1 www.who.int/influenza/gisrs_ 10 A. Jayadev, J. Stiglitz, Two Ideas To the Council of Economic Advisers and the long run. For example, only after overhauling the current IP status quo. laboratory/en/ Increase Innovation And Reduce named by Time magazine (2011) as one Pharmaceutical Costs And Prices, 2 of the 100 most influential people in the some public pressure did Gilead For too long, we have bought into https://scholarship.law.cornell.edu/clr/ Project HOPE: The People-to-People vol102/iss6/3/ world. Known for his pioneering work on rescind its “Orphan Drug”6 status for Health Foundation, Inc., 2008, the myth that today’s IP regime is www.healthaffairs.org/doi/full/10.1377/ 3 www.business-standard.com/article/ asymmetric information, Professor Stiglitz’s Remdesevir, which would have given hlthaff.28.1.w165 necessary. The proven success of health/pneumonia-kills-one-child- work focuses on income distribution, it an additional 7 years of monopoly. GISRS and other applications of every-39-seconds-127-000-died- 11 A. Jayadev, J. Stiglitz, Medicine for in-india-2018-pneumonia-cause- risk, corporate governance, public policy, “open science” shows that it is not. tomorrow: Some alternative proposals We trial vaccines, hopefully one or data-119111300489_1.html to promote socially beneficial research macroeconomics and globalization. and development in pharmaceuticals, more will generate a plausible cure, With the COVID-19 death toll rising, 4 www.axios.com/pfizer-vaccine- Macmillan Publishers Ltd, 2010, we should question the wisdom and prevnar-top-selling-drug-161f7f05-c68e- www8.gsb.columbia.edu/faculty/ and current results are promising. 4deb-93bb-c121664b7f15.html Dr. Arjun Jayadev is Associate Professor But these will also be patentable. It is morality of a system that silently jstiglitz/sites/jstiglitz/files/2010_ of Economics and Co-Director of the 5 medicinespatentpool.org Medicine_For_Tomorrow_pub.pdf sobering to note that while we think condemns millions of human beings Asian Political Economy program at the 6 to suffering and death every year. www.fiercepharma.com/pharma/ University of Massachusetts Boston and a of Salk’s polio vaccine as the archetypal gilead-asks-fda-to-rescind-remdesivir- faculty member at Azim Premji University. example of a vaccine—made freely It’s time for a new approach9. orphan-drug-tag-after-public-backlash He is a consultant to the Institute for New available, so that polio was soon brought Academics and policymakers have 7 www.thehindubusinessline.com/news/ Economic Thinking where he is Deputy under control—many vaccines are 10 covid-19-vaccine-wealthy-nations-have- already come forward with many secured-over-50-per-cent-of-promised- Director of the Political Economy of coming to market after being patented, 11 doses-says-oxfam/article32640307.ece promising proposals for generating Distribution Program. which means that originator companies socially useful—rather than 8 www.unaids.org/en/ will have the right to charge high prices merely profitable—pharmaceutical resources/presscentre/ pressreleaseandstatementarchive/2020/ Mr. Achal Prabhala is a fellow at and to decide who will obtain the innovation. There has never been may/20200514_covid19-vaccine 7 the Shuttleworth Foundation and the patent. Oxfam recently reported that a better time to start putting 9 D. Baker, A. Jayadev and J. Stiglitz, Coordinator of the AccessIBSA project, the richest countries, with 17 percent Innovation, Intellectual Property, and these ideas into practice. which campaigns for access to medicines of the world’s population, have already Development: A better set of approaches for the 21st century, accessibsa.org, in India, Brazil, and South Africa. He is known cornered over 50% of the promised 2017, https://cepr.net/images/stories/ for his work on intellectual property rights. doses available from the promising reports/baker-jayadev-stiglitz- innovation-ip-development-2017-07.pdf vaccine candidate producers. The organization estimates that the UK have obtained commitments of five doses per capita, while Bangladesh

10 COLLER VENTURE REVIEW 11 The Curious Case of Patent Balance Sheet Invisibility

John A. Squires, Esq. David N. Lawrence, Esq. Partner and Chair of Dilworth Paxson’s Founder and Chief Collaborative Officer of the Emerging Company and IP Practice. Risk Assistance Network + Exchange (RANE). Previously Chief IP Counsel for Former Associate General Counsel, Managing Goldman Sachs & Co. Director, and Head of Business Intelligence at Adjunct Professor at the University of Goldman Sachs & Co. and Assistant United Pennsylvania Carey School of Law. States Attorney in the Southern District of New York.

PO’s and M&A are back on Wall rather intangible assets – generally, Street, and a slew of tech IPO’s at intellectual property protectable as that. And while companies polish copyrights, trademarks – and, yes, And while companies I their balance sheets in hopes of patents1. Trade secret protection for capital markets success, try as you companies remains a viable option – polish their balance might a company’s patents are virtually such as the formula for Coca-Cola – sheets in hopes of invisible on corporate books. Despite yet as between patents and trade this, patents have been identified secrets, government patent programs capital markets as potentially playing a key role in incentivize patent protection, and success, try as you invariably promoting or retarding deals the markets prefer it. Patents in- subject to regulatory review in cases and-of-themselves may have value might a company’s of technologies critical to national for licensing, yet they are at their security or in response to public health most valuable when they cover a real patents are virtually crises such as by pooling. In the age of economy good or service produced invisible on transparency, wouldn’t it be important under it, allowing for government to know the economic value of what sanctioned monopoly profits. corporate books is at stake? Economically speaking as Yet, look at any balance sheet of any to patents, if you can’t measure it, you technology company lining up to go can’t control it. That is a sad statement public or even those that have been to make as to an entire class of modern long public and, with few exceptions, society’s most valuable assets. you’ll see nary a patent, let alone any disciplined accounting treatment of it. Indeed, patent invisibility is all the more puzzling when you consider In the M&A context, particularly that, since 1995, the predominant when it comes to international component of market capitalization of investment in U.S. companies that companies comprising the S&P 500 is requires approval by the Committee not the green-shade favorite of plant, on Foreign Investment in the U.S. equipment and tangible assets, but (known as “CFIUS”), lack of •

12 COLLER VENTURE REVIEW 13 accounting treatment for patents The Patent-curious Case makes the evaluation of so-called for Treatment Alternative While patent licensing ‘critical technologies’ raising national Investments security issues all the more difficult is revenue derived because it’s unclear how a company Ocean Tomo, an intellectual property values its intellectual property merchant bank, has tracked the from a company’s relative percentage of value of tangible covering those technologies. completely legal means assets (land, plant, and equipment) Commercially, without transparency to intangible assets (copyrights, of ordering the market, as to a company’s marked-to-market trademarks, and patents) over it doesn’t tell us the reports of what its patents are worth the last four decades. The results makes concerted collective action – are remarkable; the inversion of ‘price’ or the value of a such as standard setting and patent intangible assets overtaking tangible patent as an asset per pooling – all the more difficult. Indeed, as a matter of corporate value as Nobel Laurate Joseph Stiglitz occurred between 1985 and 2000. se. As a result, licensing suggests on these companion pages, Indeed, Ocean Tomo updated its study if patent pooling and ready rights revenues do not fully to account for the measurable economic access can hasten technologies to the impact of COVID-19 and found that inform or provide public in response to public health the pandemic in fact accelerated the crises, such as the current COVID-19 a price discovery trend toward intangible assets, with global pandemic, then transparency intangible assets now presenting over mechanism as the value of rights contributed to 90% of the S&P500 market value. the pool, can promote deal making. In this way, sunshine can be both the In Asian markets, however, including best disinfectant and deal-accelerant. in China, Japan, and South Korea, their observation has been a decline as Accountants would tell you that lack evidenced by the Shanghai Shenzen of GAAP2 standards is the culprit CSI 300, the Nikkei 225, and KOSDAQ conveniently enough, but to a large Composite Index, respectively. As to extent that merely begs the question. the decline, reporting differences as While Wall Street can wax eloquent Might such analysis – and even positive products they cover in complex value conundrum well over a decade ago to COVID-19 cases stemming from about strange beasts called non- collective action, such as pooling, chains. Mr. Phelps ran out of flags – in the U.S. when evaluating the various countries and difficulties in priced alternative investments, when fueled by measurable metrics such and room to insert them – at 100. so-called ‘automatic injunction’ rule economic correlation were noted. it comes to patents, accountants as the effect of the global pandemic believed to be the inexorable result While patent licensing is revenue and bankers alike are tongue-tied. on intangible assets – be aided by of a court finding of infringement. derived from a company’s completely It is time for that to change. consistent accounting treatment Justice Kennedy noted valuation legal means of ordering the market, it and balance-sheet transparency? difficulties in a famous concurrence in doesn’t tell us the ‘price’ or the value of Source: Ocean Tomo, LLC Intangible the eBay v. MercExchange case that: Asset Market Value Study, 2020 a patent as an asset per se. As a result, Components of S&P 500 Market Value *Interim study update as of 7/1/2020 Patents Un-Siloed: licensing revenues do not fully inform [i]n many cases now arising… or provide a price discovery mechanism the nature of the patent being As to patents, some companies, notably (for example, what if the patents are enforced and the economic function IBM, consistently drop billions of 100 not licensed?) as to the intrinsic value of the patent holder present dollars in revenue to the bottom line of the patent – as an asset per se. considerations quite unlike earlier from its global licensing program. 80 cases. And industry has developed As to licensing, legendary IBM, then In the financial world, ‘price discovery’ in which firms use the patents not Microsoft’s IP lawyer Marshall Phelps is normally difficult for alternative 0 as a basis for producing and selling has famously recounted the story of assets. For patents, it has been virtually goods, but, instead, primarily open innovation and licensing at IBM, non-existent and that has led to a 0 for obtaining licensing fees. whereby he informed his new CEO – speculation-laden arbitrage swamp, Lou Gerstner – that he planned to which in turn has led to the vilification In more mature and research 20 license IBM’s massive patent portfolio of entities that do not produce products development intensive industries – to the marketplace. Mr. Phelps’s team or services under their own patents such as manufacturing or 0 as ‘trolls’. This is a result where patent pharmaceuticals – patent valuation 1975 1985 1995 2005 2015 2020* then exposed an IBM laptop circuit board and inserted a flag into every holders provide no direct economic tends to bear a tighter correlation to component representing someone contribution at all – where there economic value. But in less mature Tangible assets Intangible assets else’s patent. Mr. Phelps’s story vividly are no operations directed to real and especially more ‘hard’ and ‘soft’ illustrates the interdependence economy goods and services. The U.S. technology-intensive industries, and interoperability of patents and Supreme Court noted this valuation the economic equation – even if •

14 COLLER VENTURE REVIEW 15 there is one – falters and remains Models are being developed that This, in turn, could aid in determining 1 Ocean Tomo report: virtually balance-sheet invisible more precisely national security http://www.oceantomo.com/ analogize patent rights asset as a quasi- intangible-asset-market-value-study/ as a corporate intangible asset. financial instrument in and of itself. risks as presented by investment and As to promoting desirable 2 Generally Accepted Accounting The instrument? Well, derivatives, exactly which intangible assets are If patents are to be considered an Principles: http://www.investopedia. collective behavior, such of course (this is Wall Street, after valuable and to what extent by the com/terms/g/gaap.asp asset, then they are an asset that is all) since the very essence of patent putative international investor. off-balance sheet and, non-priced, as cross-licensing or is that it derives its value from its as an “Alternative Asset,” one that patent pooling, a GAAP- Finally, as to promoting desirable does not conform to traditional asset enforceability against real-economy collective behavior, such as cross- About produced goods and services. class notions like stocks and bonds. like, consistent accounting licensing or patent pooling, a GAAP- Because alternative assets are not treatment would provide Such market-friendlier monetization like, consistent accounting treatment John A. Squires, Esq. is Partner and very liquid, valuation can be difficult. alternatives have become available would provide a market-based view Chairman of IP and Emerging Company and valuation of the patent rights Practice at Dilworth Paxson. He was a market-based view and because patent-as-derivatives can be being contributed. The market previously Chief IP counsel for Goldman An Investment Lens For valuation of the patent valued in absentia of a transaction. This approach relies on market forces confidence that would arise from Sachs & Co. He is an Adjunct Professor Patent Transparency at the University of Pennsylvania Carey rights being contributed and calculating compensation to such valuations would reduce deal School of Law where he teaches courses What makes patents ‘alternative’ in the the patent owner. The patent maps friction by promoting transparency for students studying in the International realm of financing is their nature as a and tighter correlation as to both (recall Mr. Phelps’ flags) that need to LLM program. legal property right of sorts. To enforce be created for patent valuation can rights and hard technologies pooled. a patent is to incur steep litigation costs and should highlight the correlation Consistent balance-sheet accounting David N. Lawrence, Esq. is the Founder to try your action in court and hope for between the ‘patent rights world’ treatment of patents as corporate a favorable but post-facto infringement and Chief Collaborative Officer of the Risk and the ‘real-economy’ goods and assets would increase comfort and Assistance Network & Exchange (RANE) determination by a deciding court. services world that patent claims confidence across the board, be and former Global Head of the Business If a patent holder successfully cover. If this is done right, then better it in a CFIUS review or a pooling Intelligence Group at Goldman Sachs enforces its patent, notably the legal patent ‘price discovery’ and market negotiation. It would provide current, & Co. He served for approximately 20 ‘valuation’ occurs AFTER – sometimes efficiencies will result since such market-based information that has years as Associate General Counsel and years – a trial on the merits. a mapping process ‘prices’ patents heretofore been largely guesswork. Managing Director at Goldman Sachs & Co. granularly relating to specific claim Previously, he served in various executive As a result, early notions of this time- sets’ derivative-based fundamentals positions at the United States Attorney’s warped ‘patent market’ looked and felt Office in the Southern District of New York. (that is, the real-economy value driver). like an enormous arbitrage play. That is, with the sticker-shock-high cost of This in turn would beget more patent litigation and the inherent post- consistent accounting treatment facto timing of a court outcome, patent and allow for more balance- market ‘forces’ remain untethered to sheet transparency of intangible any real economy underpinnings. assets. And this would benefit investors not only in U.S. public Currently, however, banks, private companies but also corporate equity players and hedge funds deals on the international front. have started to provide financing strategies which move beyond royalty Particularly with the U.S. and China securitizations and treat, deploy, and international agreements as to realize sustainable returns on patents intellectual property, patent balance- as assets-per se. For example, patent- sheet transparency could aid in backed loans can be structured due U.S. CFIUS review of international to a better, up-front and more ‘market mergers, acquisitions or takeovers by friendly’ valuation mechanism – enabling better determinations of the based upon credit-return models – drivers of the deal rationale. With such to become effectively more ‘liquid.’ transparency, valuation could become more standardized as to the illusive intangible assets that patents represent.

16 COLLER VENTURE REVIEW 17 Overview

ii ur Deep Innovation section frames questions 21 related to technology-led transformation. In this issue, we address artificial intelligence, and dig Artificial Intelligence: Odeep into an innovation characterized by a lack Sifting the Facts of consistent understanding about what the technology is, Dr. Aya Soffer including discussion on the potential benefits and risks. VP, AI Technology, In a spirited interview, we are joined by two leaders in IBM Global Research the field: Kartik Hosanager, the John C. Hower Professor Professor Kartik Hosanger of Technology and Digital Business at the Wharton School Professor of Marketing, of the University of Pennsylvania; and Aya Soffer, Vice Deep Innovation John C. Hower Professor of President of AI Tech at IBM Global Research. Technology and Digital Business, The section is complemented by a commentary by Lior Wharton School of the Zalmanson and Gal Oestreicher-Singer, both faculty University of Pennsylvania in Technology and Information Management at the Is AI more Artificial Coller School of Management at Tel Aviv University. Their discussion clearly underlines both the technical and management challenges of bringing ground-breaking 32 artificial intelligence applications into practice. than Intelligent? The Surprising Role that Together, our contributors clarify an important AI Plays in Management emerging technology, the advantages it promises, and the reality in practice. Dr. Lior Zalmanson Senior Lecturer of Technology and Looking forward, it seems clear that the tension between Information Management, reality and practice, between reality and future promise, Coller School of Management, are related to many technologies beyond artificial Tel Aviv University intelligence. Future versions of Deep Innovation will continue to bring together varied perspectives on such Professor Gal Oestreicher-Singer new technologies, with the aim of promoting new Professor of Technology and Information Management, syntheses and insights. Coller School of Management, Tel Aviv University

18 COLLER VENTURE REVIEW 19 Artificial Intelligence: Sifting the Facts

Artificial intelligence continues to attract strong interest from investors and entrepreneurs alike, even during the global pandemic. But where is the sector headed? Which recent developments have been most overhyped or underplayed? In this AI Roundtable, Coller Venture Review speaks with IBM’s Aya Soffer and Wharton’s Kartik Hosanagar to do a reality check.

espite the global COVID-19 Coller Venture Review discussed pandemic and economic these questions and more at a recent crisis, artificial intelligence AI Roundtable meeting with Dr. D(AI) continues to attract Aya Soffer, Vice President of AI strong interest from investors and Technology at Haifa Research Lab entrepreneurs alike. According to CB in Haifa, Israel, and Professor Kartik Insights, a New York City firm that Hosanagar, John C. Hower Professor monitors startups and venture capital, of Technology and Digital Business at while AI deals declined in the first the Wharton School of the University quarter of 2020, funding “jumped of Pennsylvania, who oversees the by 51% from the previous quarter to school’s AI for Business initiative. hit $8.4 billion.” Successful IPOs by Dr. Leslie Broudo, Head of the AI-powered startups such as the Coller Institute of Venture at insurance firm Lemonade—whose Tel Aviv University’s Coller School market cap soared to $3 billion when of Management, moderated the it went public in July—have added conversation. Following greetings more sizzle to the sector. and brief re-introductions, an edited version of the discussion appears below. As this momentum continues, several questions arise about AI and where This article is associated with a it is headed. Among them: Will AI be companion piece by Dr. Lior Zalmanson as transformational as, say, mobile and Professor Gal Oestreicher-Singer, or cloud computing? Which recent both lecturers in Technology and developments in AI have been most Information Management at the overhyped or underplayed? What Coller School of Management at challenges in AI deployment are Tel Aviv University. • unique to enterprises as compared with consumer applications?

20 COLLER VENTURE REVIEW 21 Broudo — If you think about the back end, AI or Mobile or the Internet, they AI is good at narrow tasks such as Will artificial intelligence be holds the key to a lot of what we see have been transformational reading X-rays and other such tasks as transformative as mobile that enables the miniaturization in terms of touching many in radiology. But to believe that AI is AI is good at narrow or cloud computing? of mobile devices and what we see industries and many aspects of smarter than doctors makes it seem on how the cloud runs. We use AI our personal lives. AI is similar. It like general intelligence, which it tasks such as reading Soffer — to make sure the cloud is running is advancing so rapidly that there is not. That, to me, is the biggest I think AI will be as X-rays and other such smoothly; we use AI to predict is little doubt in my mind that it is myth that needs to be exploded. transformational as mobile and failures and fix them in advance. significantly transformational. tasks in radiology. cloud computing but maybe in Soffer — In my view, AI touches almost a slightly different way. That is Broudo — I whole-heartedly agree with everything we do from the core of But to believe that AI because AI touches everything. To summarize, from both your Kartik. This is a question I get a creating the technology all the way is smarter than doctors Part of the transformation in perspectives—the technology- lot. People are so anxious about AI. to user interfaces. In the future AI cloud and mobile—and enabling enabling component and the change Will it make decisions for us and will be even more transformative makes it seem like all that—is due to AI. AI is the in industries it is affecting—you take over? People get these ideas when it will allow everyone—even general intelligence, underlying capability that has made believe that AI is highly significant. when they read the popular press. those without computer skills— all these things transformative When you read the popular press Then you go to the lab and see the to interact with technology. which it is not and will continue to do so even coverage of AI, what do you think is gap between what they believe Kartik Hosanagar more moving forward. Hosanagar — being overhyped? And what is being to be close to general intelligence If we rewind back the last 20 to 25 downplayed to the degree that we and where we really are. That If you consider mobile, there are years, we can look at technologies do not appreciate what is going on? is what I think is overhyped. things we like and others that we that have had a huge impact on may not like as much but are very Hosanagar — business and society, such as the Recently an idea that has received If we look at the popular press, Internet, the Cloud, and Mobile a lot of coverage in the popular I believe where there is lack of and being able to reason or reach computing. There also have been press concerns a text generator— understanding is the view that AI conclusions in a meaningful way. other technologies that may have GPT-3 (generative pre-trained is almost magical and sentient, and It is confusing, in that sense. received a lot of hype but which transformer), as it is called. The it can be viewed in the same way as have failed to deliver. To me, AI risks of this technology relate to Broudo — human intelligence. There is also belongs in the first bucket with the what has come to be known as To turn now to the second half this concept of super intelligence. In Internet and Mobile computing. deepfakes. These things, on the of the question, what are we not reality, AI today is what we can refer surface, can seem very real—and seeing? What do you think is When we ask whether AI belongs to as weak AI or artificial narrow that is why they are overhyped around the corner, that people are in one bucket or the other, I think intelligence. That means we give a and get a lot of attention. It seems not seeing yet and is getting lost? about it in a couple of ways. First, machine learning algorithm good like a computer can write a poem Hosanagar — how relevant is the technology data on one specific task and we can or a news article. There have For me, that would be the fact that across multiple industries or figure out the patterns that allow been a lot of articles about that. modes of our life. On this count, us to make predictions on that task. when you look at AI, in practice, AI is fundamental. We see AI For example, we can give data on When you look at what the 90% of it is machine learning—and applications in health care, finance, whether an email is spam or not, and computer is doing, you can see 90% of that is supervised machine education, professional services, it can do a great job of figuring that that it is not truly intelligent as learning. That is the idea that you manufacturing, retail and so on. out. But that does not mean it can we think of intelligence. But it is have massive data sets which relate The scope is extremely broad. be truly intelligent and transform getting there, and it raises what to what you are trying to predict. You learn from those data sets so that Second, over the last 20 years we itself into a robot that starts moving I think are interesting questions you can start making predictions. have seen an explosion of data in about in the physical world. You about what intelligence is. Anybody One of the exciting things that all aspects of our lives. We could might have an autonomous vehicle who understands the technology could happen is that there are other not have mined that data 20 or 25 that drives around, but that does not behind it knows that it has ingested necessary. One reason that mobile approaches as well. One of them, years ago. Now, we not only have mean it can do other things. That so much information that no matter has taken off more than we had for example, is reinforcement data generation happening at an is where we are. what you ask, it can come up with envisioned is because we can now learning. That is the idea that an AI amazingly fast clip, but we have also what seems like a reasonable AI touches almost know people’s locations. That is AI Sometimes, when you read articles system observes what happens and seen machine learning progress answer. That happens until you in the mobile environment. That is in the popular media, you start to see learns from it—and you do not need everything we do from so much that we can analyze start asking it things it does not why “mobile” technology can now the discussion around AI suggesting massive data sets to do that. The that data and make sense of it. know. That is when you see that the core of creating the that we have created something idea that you can create intelligence provide users with an experience its response is nonsensical. technology all the way that is personal. Even though A lot of factors have come into that is truly intelligent and mimics without massive training data sometimes we are surprised (or play at the same time, ranging human intelligence. We are nowhere I do think something interesting is sets is interesting. Reinforcement to user interfaces possibly dismayed) at how much from data generation to data near that. We might get there in happening, and a lot of it is exciting. learning has been around for a long Aya Soffer these technologies know about us, processing to progress in machine the future but we are not there yet. But on the other hand, it is far from time—but in terms of industry that is what makes them extremely learning algorithms to the fact Similarly, there is a misperception what we would consider intelligence applications and business settings useful. That is on the front end. that this is happening across that AI can beat doctors or trump in the sense of understanding in which it can be used, those industries. If you look at Cloud their medical knowledge. Here, too, the rules or laws of the world are not as well understood. •

22 COLLER VENTURE REVIEW 23 To me, the future developments that look promising are in what people are calling neuro-symbolic AI… That is what will help us go to the next level of being able to understand language as well as rationalize and reason. Adding knowledge and rules is where we want to go Aya Soffer

Supervised machine learning is This was because, as Kartik has Broudo — focused on its learning from the past, pointed out, it dealt with narrow What challenges in AI and where lots of historical data exists. tasks where, with enough examples, its deployment are unique to In contrast, reinforcement learning we could predict or classify things enterprises as opposed to AI in is about acting after observing what in a fairly good way. However, consumer applications? What happens, and it allows us to learn the issue we have today is that critical differences should we from the past without training it is impossible to machine- be aware of when we think data. For example, imagine drug learn everything. about AI in the B2B model discovery. You could apply this First, this is because we do not versus the consumer model? concept to find drug molecules in have all the data; and second, Soffer — situations where you do not have because some things are much We at IBM work predominantly much training data. There are many too intricate. This is also how we in the B2B model, so I will tell you interesting applications. The idea learn as humans. We learn some what I have observed. The first that you can learn even in situations things through patterns: If you see thing to note is that B2B is B2B2C where we do not have a lot of data enough images of cats, you know eventually. When you deal with from the past is super interesting. what a cat is. Second, we learn other businesses, you realize they have Soffer — things because we go to school, many issues around deploying AI. To me, the future developments and somebody explains them to They care a lot, of course, about the that look promising are in what us. If you are taught that a cat goes outcome but they also care about “meow,” you know that a “meow” people are calling neuro-symbolic the infrastructure. For example, sound means a cat. Where we need AI. If we look at the history of questions can come up, such as, to go now is to combine these two. AI, originally the concept was “How much it will cost to run that we were somehow going to The neural architecture and AI?” or “Will those costs create codify all the knowledge in the machine learning can help us to enough value compared to the world. We had expert systems, learn better. We can use neural alternative of not running AI?” we had logic reasoners, but that networks to perform logical A lot of economic considerations never really took off because it reasoning on the data in the come up because running AI not is impossible to codify all the knowledge bases. That is what only requires a lot of data but also world’s knowledge and rules. will help us go to the next level computing power. It is important Plus, it does not scale because of being able to understand to understand the KPIs or key it is hard to prove these things. language as well as rationalize and performance indicators around AI. reason. Adding knowledge and It must deliver outcomes that Instead, along came the paradigm rules is where we want to go. of machine learning, which has are accurate, and it is equally turned out to be highly successful. important that AI must help a •

24 COLLER VENTURE REVIEW 25 life? That issue has not received behavior that seems intelligent. as much attention as it deserves. If we look at human beings, what differentiates us from others and In fact, several studies in the social An interesting question what makes us intelligent is the sciences and psychology show that that often comes up is what ability to develop human language people tend to have some algorithm to communicate. That is part of aversion, especially when they see it will take for that user to everything we do—whether we an algorithm fail. And no AI is a trust AI. Would a doctor be are talking with our family at home, perfect system. It might on average in school getting an education, be better than human performance, willing to trust the judgment or at work. With a machine, that but an AI system can go wrong. How of a diagnostic AI or would is not the case. I believe that as will people react when it fails? end-consumers be willing the field of AI matures, it will help Aya brought up explainability, but machines to communicate with if you don’t even understand the to trust a prediction from people using plain language. That system and you have seen it fail will create the ability for every company achieve its goals. The borrower has been denied a loan an AI system and apply it and you continue using it, what single person on earth to use goals can range from functional [based on an AI recommendation], will it take for such an AI system to their life? How will people computers and language. Humans ones—such as increasing sales— we must be able to explain why the to be adopted? Do we know if it can communicate through language to non-functional ones, like the loan was turned down. Companies react when it fails? is performing universally well for by the time they are two years old, necessity to invest millions of are not happy with black-box AIs. all people? Maybe it is beating Kartik Hosanagar but computers still cannot do that. We may develop more and more dollars in computer farms. Those lay people, but it is not beating sophisticated black box neural The question that arises is, why are some of the practical aspects. experts. Some questions that relate networks, but people tell us, “No, has that been so hard? Where Such considerations make AI to human psychology will start please bring me back my rules are we on the journey of trying to matter a lot when it comes to deployment different than software and things that I can explain to to do better? Anyone who has designing the interface between engineering. How to deploy new people.” There is friction between How to deploy new used personal assistants [such as humans and AI. It will also matter software is relatively easy and now what businesses want and feel Alexa, Siri, etc.] or chatbots can software is relatively what information is shared with well understood; with AI, that is not comfortable with, versus the way AI understand that those programs are humans and what information is yet well understood. We have this really works. That makes it harder easy and now well codified. They provide responses withheld to make it easy for them to notion of AI life-cycle management. to deploy AI at the enterprise level. based on simple questions. understood; with understand and encourage adoption. You train your models on data, then That is where we need to go. Hosanagar — you test them—though no one knows AI, that is not yet Broudo — Aya mentioned that B2B is often Technology in recent years—like yet how to test AI the way you test We turn now to natural language B2B2C. I would like to add a point well understood GPT-3, which I mentioned before, software. Unlike software, the processing (NLP). Where is the field from the end-user standpoint— which again stands for generative performance changes if the data Aya Soffer headed in the next couple of years? whether that user is sitting in an pre-trained transformer—has changes. In software engineering, Why does it stand out in your mind enterprise or at home. An interesting taken a big step forward. It can the software does not change with as an area we should care about? question that often comes up is help with understanding words the data. But in AI, if you have new what it will take for that user to trust Soffer — that are used in a certain context, data with other statistics, your models AI. Would a doctor be willing to Let me begin with the second but it does not help with the may no longer be relevant. These are trust the judgment of a diagnostic question about why we should nuances of the language. That some of the issues around deployment. AI or would end-consumers be care about NLP. Most of AI today takes us back to the challenge of Another area I would highlight is willing to trust a prediction from is machine learning and analytics, reasoning. That is what we will explainability. For example, if a an AI system and apply it to their but AI is about machines exhibiting have to do to crack language. •

26 COLLER VENTURE REVIEW 27 I do not think you can learn a I believe we will see more maturity building? If so, what kind of language just by seeing more in the market. Some investments companies are you looking for? and more examples [of how it is are being made in pure-play AI Soffer — used]. Reasoning in a meaningful startups—but those are becoming Generally speaking, many startups way cannot be learned simply by increasingly hard. Large players are building horizontal capabilities crunching more and more data, like Microsoft, Amazon and others or tooling for the AI world. I agree but that will be necessary before are creating AI-enabled tools and with Kartik that they will find it hard computers can converse with giving them away almost for free to remain independent and grow us. Ultimately, natural language as part of their cloud infrastructure. that business. Companies want interfaces will do everything we A startup that comes in with a to run their AI on the cloud or the do and completely change the horizontal AI application that hybrid cloud, which is a combination way we interact with computers. can be applied across many of on-premises (or private) cloud For example, if you are a doctor, industries will find it increasingly and the public cloud. On the other the computer should be able to difficult [to generate revenues]. hand, there are many ways a small recommend the next thing to If Google or Amazon comes along to ask the algorithm to write about Hosanagar — company can innovate faster than do to help the patient. And it something new­—such as COVID-19, AI is a big buzzword in the startup with a free AI product as part of a big company can. I do believe will also have explainability; the before the pandemic was out there— world. If a startup claims to have its cloud infrastructure, then it these smaller companies will be algorithm will be able to explain Ultimately, natural becomes extremely challenging it would not be able to write about AI, that bumps up its valuation absorbed eventually by some of itself. It is all about understanding for a startup in that space. something it did not know. It would language interfaces and increases the chances that the larger ones. Eventually these and generating language. not be able to write about a new capabilities may become part of the Like AI, it will impact many will do everything we the venture will be funded. As a As a result, while there are vaccine, for example. It cannot do big platforms. The startups that will applications in many industries. result, lots of startups claim to have some startups in the horizontal that. You would have to reprogram do and completely AI. When I said 90% of AI today is space, more are coming up in the become large in AI will be those that Hosanagar — it with many, many articles about change the way we machine learning, I should have vertical space and bringing AI focus on specific industry use cases. I agree with everything Aya said, COVID-19 before it could write clarified and said that 90% of real to deal with a specific problem. Broudo — including what she said earlier an article about COVID-19. interact with computers AI is machine learning. The truth For example, it could be AI for a Finally, as a last question: Investors about GPT-3 and the progress being One strong feature of GPT-3 is that is that 90% of what passes for AI is personalized medicine application can throw their money in places made. It is an extremely exciting Aya Soffer it has been able to pick up style. That not really AI—people claim that or for fraud detection in the credit that change our world but one could space—and one in which we will is why people are getting excited everything is AI if it touches data card industry. We see a lot of that say there is no clear governing see a lot over the next few years. about it. But several articles about A problem with these systems even slightly. A lot of that is going on. kind of activity. That is a little authority. What should guide us in more defensible for the startup. the absence of global standard rules? Broudo — GPT-3 have also pointed out its sometimes is that each sentence That said, however, investors are In September The Guardian deficiencies. Often the words make may make sense, but the paragraph starting to get savvy about AI. Some Broudo — Soffer — published an opinion piece titled, sense, but the content does not may not make sense. At the same AI-specific venture funds have been Aya, is IBM buying some of Education is important; I would “A Robot Wrote This Entire Article. always make sense. It is written in time, if you know what you are created; some of them invest only these startups? Do you see augment that with transparency. Are You Scared Yet, Human?” exceptionally good English, and it doing, you may be able to fine- in AI startups. As this happens, value in investing rather than Regulation may help AI to become The newspaper claimed that the sounds like something that a highly tune GPT-3, give it very specific more transparent so people can article—which was generated educated person might write, but training data, and then tune it make better decisions. In IBM, by GPT-3 — took less time to edit that doesn’t mean the content itself to produce the kind of result you specifically, we are doing our best to than many human op-eds. Is that makes sense. It is interesting, and want. If you think in a very narrow, pursue an idea—in partnership with why there has been such a buzz we will see where it goes. On the targeted manner, you may be able other companies, of course—that over GPT-3 compared with other syntax, GPT-3 is excellent. On the to get it to do the kind of writing we call AI FactSheets. If you think instances of natural language semantics, it is still lacking. Still, it that seems almost human. about nutrition labels for food, those generation? Where are the limits is a powerful NLP tool that will help Broudo — were not there in the beginning. to what these technologies can do? us build better systems in the future. Kartik, since you are in the Over time, more and more Soffer — Hosanagar — entrepreneurship space, how regulations dictated that companies The reason people are paying more In practice, creativity often means do you view investors’ approach had to display labels on food attention to GPT-3—compared to combining things in interesting to AI? Does every deal have to packaging so that people could know the previous version GPT-2, which ways. We should not assume that have an AI component? If so, what they were eating or drinking. was also amazing—is because of these systems are unable to come does it drive a higher valuation? Similarly, with AI factsheets, the amount of data it was trained up with things that are novel and How well does the startup world we will have a form that says on. Its full version has a capacity even creative. They will combine understand AI, and is that different you need to describe your model, of more than 175 billion machine things in interesting ways. The than what you have seen before? how you trained it, how accurate learning parameters. As a result, it scope of their creativity may be it is, and things along those lines. can generate language that on the limited, but that is not in the same That transparency, which is surface seems very natural. That way that a lay person looking at it something that can be regulated, is the reason why it has received would rate it as being creative or not. will let people know what is healthy so much attention. But if you were and what is unhealthy in the •

28 COLLER VENTURE REVIEW 29 I do believe we need governance standards. The industry is Governments must start participating in forums such as Partnership on AI to discuss best getting savvy about how practices. My observation is that AI can be regulated without scientists from leading companies are coming together to discuss how stifling what is innovative to use AI responsibly. Still, when push comes to shove and decisions Kartik Hosanagar are made higher up in these organizations with a view to meeting quarterly targets, some of these conversations might not matter. The governance frameworks consumption of AI—just as they should not be limited to companies About About the Author do today with food consumption— self-regulating, in my opinion.

so that they can make better and Governments must start getting Dr. Aya Soffer is VP of AI Technologies for Professor Kartik Hosanagar is the John C. more informed decisions. savvy about how AI can be regulated the IBM Research AI organization focusing Hower Professor of Technology and Digital without stifling what is innovative. Hosanagar — on natural language understanding and Business and a Professor of Marketing at This will require participation by We spoke about how AI can be conversational systems and their application the Wharton School of the University of consumers, who will need to be transformative, and it is progressing in customer care and other enterprise Pennsylvania. His research focuses on the educated about the technology and applications. In this role Dr. Soffer is digital economy, in particular the impact at a rapid pace. We also discussed its risks. AI should be part of the responsible for setting the strategy and of analytics and algorithms on consumers how we are in a world of artificial curriculum in schools, so people working with IBM scientists around the and society, Internet media, Internet narrow intelligence and are inching understand what AI is and what it world to shape their ideas into new marketing and e-commerce. Professor towards general intelligence. We AI technology, and with IBM’s product Hosanagar serves as a department editor can and cannot do. For example, if are going to have situations where groups and customers to drive research of Management Science and has previously they apply for a loan, they should AI can be used for good, but it innovation into the market. served as a Senior Editor of Information can also be used irresponsibly. know what assumptions have been Systems Research and MIS Quarterly. built into an automated system. In her 20 years at IBM, Dr. Soffer has led Aya mentioned technologies such Or, if they read an article online, several strategic initiatives that grew into Professor Hosanagar cofounded and successful IBM products and solutions in developed the core IP for Yodle Inc, a as GPT-3 and the fact that in the they should know how to assess the Big Data and AI space including the venture-backed firm that was acquired wrong hands, fake news articles whether the information they can be produced at scale without original Watson system and more recently by Web.com. are consuming is truthful. Most Project Debater. She has authored over any human beings being required importantly, education will need 50 peer-reviewed papers and served as an to generate them. Photos and videos to change so that people know invited speaker in numerous conferences. can be doctored to produce deep how to function in a world where fakes. There is also the business AI is an active participant. of using AI for loan approvals or in the judicial system to make parole decisions. We do not need someone to have nefarious intentions for things to go wrong. All that is required is a slight oversight— and you may end up with a biased algorithm that makes discriminatory loan decisions that impact millions of people. It is not that humans are not biased; I do believe that, on average, AI will be less biased than humans. A biased judge might affect the lives of 200 or 300 people; a biased HR manager may make poor decisions about a few thousand people; but if an AI system is deployed to make decisions at scale, bias in those decisions may impact millions of people.

30 COLLER VENTURE REVIEW 31 The Surprising Role that AI Plays in Management

Dr. Lior Zalmanson Professor Gal Oestreicher- Singer Senior Lecturer of Technology and Professor of Technology and Information Management, Information Management, Coller School of Management, Coller School of Management, Tel Aviv University Tel Aviv University

uppose we are to form our other direct bosses and officially AI business strategy based are not even considered employees on how Artificial Intelligence but rather freelancers. In reality, Sis being portrayed in popular however, they are being managed by media. In that case, we will probably artificial intelligence algorithms. be limited to one of the two common When AI algorithms become “your notions. The first is AI as a servant, boss,” new tensions emerge. Drivers embodied in the human environment experience tensions related to the through robotics, helping humans in manner they conduct work since, on their daily needs. The second is AI the one hand, they are autonomous as a superintelligence, a replacement agents who choose to work at will. • for any human, an all-knowing being, controlling and overseeing anything and everything. However, the notions and roles that AI will probably play in our society, as discussed by Aya and Kartik in the Roundtable, are very different, and in a way, much more interesting. Research conducted at the Coller School of Management at Tel Aviv University might shed further light on the matter, particularly as As COVID-19 provides a catalyst related to a set of practices also known as “Algorithmic Management.” In for remote work, many firms will this type of management, algorithms have to decide how they control take over the traditional roles of middle management. This term doesn’t work from afar. It is likely that we represent a futuristic scenario. For would see different implementations Uber drivers, for example, this is very much a current reality. Such of AI algorithms taking middle drivers work under tight supervision management’s traditional roles by a machine learning algorithm, guiding their actions and sanctioning them if they do not follow the firm’s policy. They do not have

32 COLLER VENTURE REVIEW 33 guidance, mentoring, and rapport he brought participants to a class between humans is not likely to be and gave them simple perceptual replaced soon by machines. In ride- There are great tasks. When participants were alone, hailing, drivers seem desperate for they gave correct answers quickly. voice support, precisely when they possible benefits of Still, when he added other “fellow run into tension-inducing situations human-AI collaboration participants” who cited wrong answers that the algorithm cannot solve. In aloud, many participants confirmed those cases, drivers appreciated for optimal decision the majority decision and shared About the fact that the firm has built a 24-7 making; however, if the same erroneous responses. human-led support line for them. humans conform to Our research finds that the same Dr. Lior Zalmanson is a Senior Lecturer of It is important to note that algorithmic phenomenon is at play in the encounter Technology and Information Management management isn’t restricted to these AI decisions without between a human worker and an AI in the Coller School of Management at new gig workers. As COVID-19 agent. In this case, the presentation of Tel Aviv University. provides a catalyst for remote work, exercising their an AI’s advice changed the worker’s Dr. Zalmanson’s research focuses on many firms will have to decide how judgment, the results answer in a statistically significant online engagement and commitment, they control work from afar. It is could be anywhere number of cases (15-25% compared the sharing economy, algorithmic likely that we would see different to always answering correctly in the management, and the future of work. implementations of AI algorithms between sub-optimal control group). When we presented His award-winning research has been taking middle management’s to plain dangerous them with multiple AI agents, all citing recognized by and received grants traditional roles. Therefore, the the wrong advice, the percentage grew from the Fulbright Foundation, Google, tensions observed in the Uber even higher. These findings provide Marketing Science Institute, Dan David drivers’ research are likely to be Prize, and has been featured in Fast On the other hand, they are being a warning sign regarding the design expected in these future scenarios. Company and across Israeli media. surveyed and micromanaged by of human-AI hybrid decision-making pervasive technology. Drivers enjoy However, even if many firms won’t processes and calls for better work Previously, Dr. Zalmanson was a research fellow at the Metropolitan Museum Media the reliability of AI algorithms that adopt AI as bosses, they might install processes. There are great possible Lab and a visiting Assistant Professor at constantly match them with riders them in the role of non-human benefits of human-AI collaboration for NYU Stern. but at the same time feel frustrated workmates. In their research, Erik optimal decision making; however, from the lack of transparency of the Brynjolfsson and his colleagues at MIT if humans conform to AI decisions complex algorithmic calculations note that most current occupations without exercising their judgment, the Professor Gal Oestreicher-Singer is a which are in charge of their wages. won’t be replaced by AI (or specifically, results could be anywhere between Professor of Technology and Information Management in the Coller School of Working under algorithms means as Kartik and Aya mentioned, machine sub-optimal to plain dangerous. Management at Tel Aviv University. personalized treatment and a lack of learning) but instead the augmented In this aspect, we agree with Kartik’s solidarity as any worker is being treated and re-engineered by the introduction notion that “AI can be used for good, Professor Oestreicher-Singer’s research focuses on the effects of social media, differently based on their unique of such capabilities. Humans and AI but it can also be used irresponsibly.” consumer engagement and peer influence case history. In the end, many drivers will not work as substitutes but rather Behind the words “use” in this case on electronic commerce. She is a recipient reported feeling isolated and “robot- complement each other’s weaknesses. lies more than AI’s purpose and work of the prestigious Sandy Slaughter Early like.” They resorted to ad-hoc online Thus, the burning question is how context. Putting AI to good use means Career Award and of the European communities to socialize and try to to design, engineer, and manage designing responsible and transparent Research Council Grant in 2018. She make some sense of these algorithms these new human-AI work hybrids. AI processes with humans in mind. serves on the editorial boards of MIS and their behavior. In some cases, In an ongoing research project, Quarterly, Information Systems Research drivers even go further and choose to Lior presented in the international and Management Science. reject and revolt against the algorithms conference of information Systems by blocking or gaming them. (together with a Ph.D. student, Yotam Thus, a firm that chooses to manage Liel), we study the risk of humans by AI algorithms shouldn’t rush to blindly conforming to the algorithms’ take the human element out of the decisions without properly weighing equation. Over the 20th century, we them against their better judgment. learned the importance of investing The paper follows Salomon Asch’s in human resources. The support, seminal conformity research in which

34 COLLER VENTURE REVIEW 35 Overview

iii ur Virtual Roundtable brings together global 39 leaders and thinkers from top venture capital funds to address an area of significant change How the COVID-19 Crisis Oin venture, innovation, and entrepreneurship. Ignited and Accelerated In this issue, we examine the early stage funding Venture Capital ecosystem, and the ways in which leading funds are meeting the imperative to adapt and are being transformed. Fiona Darmon General Partner, In this discussion, we bring together leaders from Jerusalem Venture Partners Jerusalem Venture Partners, Bessemer Partners, and PrimeTime Partners. Fiona Darmon (Jerusalem Venture Felda Hardymon Partners) and Felda Hardymon (Bessemer Partners) are Virtual Roundtable Partner, Bessemer Venture Partners Professor of Management Practice, long-time investors who share their views in weathering Harvard Business School through another storm to bring forward the next round of startup innovation. Abby Levy, leading a new fund Shai Bernstein with veteran investor Alan Patricof, adds valuable insights The COVID-19 Crisis Associate Professor in on the opportunity for startup capital to chart a new Entrepreneurial Management, investment niche in the senior market, led by a team Harvard Business School of diversified talent. Each of these individuals’ perspectives, responding to for VC – Death Knell what is both specific and general in the changing economic and social context, helps us to consider the profound ways 46 in which the theory practice of entrepreneurship and A New VC Fund Rides innovation are informing one another. Looking forward, or Springboard? future discussions in the Roundtable section will continue the Silver Tsunami to bring together partners and collaborators active in Abby Levy forging our new venture ecosystem. Managing Partner & Co-Founder, PrimeTime Partners

36 COLLER VENTURE REVIEW 37 How the COVID-19 Crisis Ignited and Accelerated Venture Capital

Fiona Darmon Shai Bernstein General Partner, Associate Professor in Entrepreneurial Jerusalem Venture Partners Management, Harvard Business School

Felda Hardymon Partner, Bessemer Venture Partners Professor of Management Practice, Harvard Business School

Based on the experience of the financial crisis of 2008, many feared that last year’s coronavirus pandemic might devastate the world of venture capital. Instead, the industry is booming and changing, according to Harvard’s Shai Bernstein, Bessemer Venture Partners’ Felda Hardymon, and Fiona Darmon of Jerusalem Venture Partners. The three experts participated in a virtual venture capital round table organized by Coller Venture Review. •

38 COLLER VENTURE REVIEW 39 In March, at the onset of COVID, she and her colleagues at JVP “went into the trenches to strengthen the portfolio. We thought that no matter where the next 18 months were going to take the portfolio, we need to get ready. Our mind-set was, ‘No matter the uncertainty, cash is king, and this is going to be a time for opportunity.’” Their top priority was to ensure that their portfolio companies had enough cash and credit to last at least 18 months. “We secured over $200 million of cash and about $70 million of credit facilities for the companies,” hen the COVID-19 Bessemer Venture Partners had she said. “It was all about ensuring pandemic struck in approved 15 deals in two weekly that they had the resources to make early 2020, countries meetings at the end of 2020. “Several it through, and even take advantage, around the world went of those were follow-ons,” he said. W of the storm.” into lockdown, unemployment soared, “At least five were raising more than and stock markets initially tanked in $100 million at a valuation of more As the months rolled by, Darmon and March and April. It seemed at that time than $1 billion. That used to be a her colleagues realized that several that the double whammy of the health public offering. Now it’s just another companies, especially those engaged in crisis coupled with the economic shock Monday.” The influx of outside capital enterprise, data, cyber and e-commerce, might devastate the world of venture had added to the stability, he added. were thriving rather than struggling. capital investment. Much to the “The move to home offices is only just surprise of those who had expected a one major example,” she said. More replay of the Great Recession of 2008, Entrepreneurs Without Borders and more people self-isolated and that did not happen. Felda Hardymon, practiced social distancing because of a partner with global venture firm The COVID-19 crisis also accelerated the pandemic, and this sparked a surge barriers to diversity and gender Bessemer Venture Partners, says another trend—the seeming of what Darmon described as “solace balance in the VC industry. “COVID-19, in fact, ignited venture disappearance of geographic borders buying… from the comfort of your capital. The truth is, it accelerated Bernstein began with an open-ended COVID-19, in fact ignited for online business. “Entrepreneurship own room.” She sits on the board of a some parts of the industry and question: “How do you view the role venture capital. The truth is, is borderless,” Hardymon said. “Silicon company that went public last year, and brought in even more outside capital.” of venture capital overall in society as Valley doesn’t have a monopoly on it had to “digitize its packaging because These remarks were made at a virtual an asset class?” he asked. In response, it accelerated some parts of innovation anymore. The COVID everybody wanted more efficiently VC round table discussion organized Hardymon noted that the pandemic the industry and brought in [pandemic] has accelerated online sized boxes for shipments. It became in December by Tel Aviv University’s has exacerbated a trend that has existed business and brought it forward a crazy year. We have seen businesses Coller Venture Review. Shai Bernstein, for the last four or five years and even more outside capital [possibly] five to ten years.” For example, go on steroids. We were getting ready an Associate Professor of Business which has now taken off. “Growth he pointed to the software business, to brace like [the recession of] 2008, Administration at Harvard Business equity is now increasingly treated as which is migrating increasingly to the and suddenly all these categories blew School, hosted the round table with an intermediate, almost independent cloud. “We are the largest investor in out of the park.” asset class,” he said. “It has attracted cloud software in the world,” he added. Fiona Darmon, a General Partner at Among the additional sectors that hedge funds, private equity funds, “What does that mean? It means that Jerusalem Venture Partners (JVP), saw robust growth were digital health, even limited partners investing along Shopify, for example, can service the and Felda Hardymon of Bessemer telemedicine, and logistics, Darmon with venture capitalists.” The reason, globe as far as startup companies go. Venture Partners. noted. Banking was another sector Hardymon explained, is that “there is a It does not make any difference where Bernstein focused on three themes. that was forced to go digital almost perceived stability in startup companies you are. You can build a world-class First, he explored the impact of the overnight. “The banking sector in that are well-sponsored and well- company now on the cloud from any coronavirus pandemic on venture Israel didn’t even know what online managed. Having been in the business geography.” As a result of these factors, capital, especially on the sourcing and was until the pandemic kicked in,” she for 42 years, [I know that] they are not Hardymon said, “venture capital is geographic dispersion of investments. said. “Suddenly you did not have to go all well-managed; in fact, often they probably more important as an asset Second, he spoke about the emergence to your branch any more to sign the are not. But in general, there is a kind class than ever before.” of mega-deals and the concentration forms; all that moved online. Many of stability that is seen, and that has of capital, including the fact that “80% Darmon agreed with Hardymon’s mom-and-pop stores moved online. changed the way we do business.” of capital in the venture industry was view. “Both here in Israel and our Sadly, what stayed behind were the allocated to rounds more than $25 Thanks in part to the infusion of offices in NYC, everything down to small domestic businesses that couldn’t million.” Finally, he highlighted the fresh capital, Hardymon noted that the last bit resonated true,” she said. digitize their operations fast enough.” •

40 COLLER VENTURE REVIEW 41 Darmon, like Hardymon, witnessed the And coming up underneath is a new this have made VCs even somewhat waves of new capital pouring into the set of semiconductor companies, which narrower in terms of the type of industry. “We are seeing a tremendous are going to be fundamentally different companies they are willing to fund?” amount of co-investment,” she said. because we no longer depend on he asked. “If you are an early-stage “That was already the trend in 2019, but Moore’s Law.” Hardymon added investor, you need to find companies in 2020 it massively kicked up speed.” that “materials are still really, really that have the potential to absorb these interesting,” and that “some of the large sums of capital down the road.” Darmon reiterated the critical semiconductor companies are Darmon noted that the question relates importance of venture capital as an now pushing materials. Quantum asset class. “There are so many great computing is fundamentally a to a discussion going on within the deals out there now,” she noted. materials thing. We have positions in venture capital sector about the changing “Nobody is looking at the capital all of those…they are now close enough role of seed capital. “Statistics show that market as the exit scenario.” Though that they are out of the science stage in 2020 seed [investment] went down Darmon had taken a company public and into the development stage.” materially. This is primarily because a in Tel Aviv two weeks before this lot of capital has moved along the food Bernstein wondered what Darmon discussion, she did not view it as an chain. It is not necessarily because of and Hardymon thought about the trend exit. “It is a form of growth equity, and concentration, but it is just that there is of capital growth and increasingly we continue to build up these players,” a tremendous amount of money available larger rounds of VC investment. “That she said. “Everything has changed. now, and that is probably one of the triggers concentration on the venture The most important thing is that the more interesting alternatives to the side as funds are becoming bigger end game is different. We are looking capital markets to a degree.” She added to keep up with the pace. But can bigger; we are seeing bigger.” Darmon that as more capital has become available, observes more patience among Israeli VCs have been making some of it entrepreneurs because they are ready available to their portfolio companies to go further. “In terms of the role of to pursue strategic acquisitions. She the asset class, it is the most interesting did add though, that seed investments one. It is driving the economy. must continue, as these deals feed Everything has It is creating jobs. It is creating an the food chain, and create the major alternative to the old world.” companies of tomorrow, emphasizing changed. The most the need to continue to allocate a Bernstein asked Hardymon which certain number of investments to seed. sectors of industry were changing most important thing is because of the pandemic. The Bernstein asked how the presence response: “One in every ten dollars in of bigger funds—combined with the that the end game venture capital this year is going into change in early-stage investing— digital healthcare.” While that might has altered the investment strategy is different. We are seem predictable, Hardymon believes of venture capital firms. Hardymon that these investments are not a replied that it has changed the whole bubble. “It’s going to stay,” he said. VC industry. “I do think what Fiona looking bigger; we “Like so many other things, things are (Darmon) said is one way that has getting moved forward that will be that happened. An even bigger force has been just the rush into the intermediate are seeing bigger way forever.” Software as a service and growth stage capital from civilians (SaaS) is also an area that is attracting and non-venture capitalists…The way capital. “Even now, it accounts for 40% the industry invests has changed.” of our dollars going out,” he added. During the tech bubble of the late Boom in Semiconductors 1990s, the venture capital industry stage-shifted its strategy downwards. Hardymon believes “we are on the In contrast, the present transformation verge of another big boom in is different. “Two things have semiconductors.” He has a ringside happened,” Hardymon noted. “One is view of this sector as a board member that funds that can raise the money of a semiconductor company; and have done so and started to participate Bessemer has four or five semiconductor more in the middle and late rounds. companies in its portfolio. “We have You have the phenomenon of a lot of seen Nvidia buying everything that the big managers, us included, who AMD or Intel can’t buy, he said. “There have multiple funds that include has been big consolidation at the top. later-stage funds. We have also made •

42 COLLER VENTURE REVIEW 43 links down to the seed stage. We “What suddenly happens when it’s so have several seed initiatives. We are balanced, it suddenly becomes more not the only venture capital firm that natural when you’re recruiting to has that.” Some firms have alliances be looking at both women and men with entrepreneurship accelerators without realizing or consciously doing such as Tech Stars. “We have changed that.” Darmon noted that women with the way we do our business to catch strong positions in a VC firm usually things coming out and to get started cannot be brought from outside; they early, and then having the drive must be grown from within. “We have power to ride them out. As a firm— never made diversity a conscious agenda and as an industry—we look at We are 50% women [...] at JVP. It is just the way it is. There are things differently.” What suddenly happens when times when there are more women than men around the table. It is so Diversity and Gender Balance it’s so balanced, it suddenly balanced because no one feels that As the discussion neared its end, becomes a lot more natural they are there for any reason except Bernstein brought up the issue of for merit.” About About the Author About the Author diversity and gender balance in the when you’re recruiting to be Bernstein concluded the round table by venture capital field. According to Fiona Darmon, General Felda Hardymon, Partner at Professor Shai Bernstein looking at both women and asking both participants what should Hardymon, this is an important issue, Partner at Jerusalem Bessemer Venture Partners is Associate Professor be done to make the venture capital and “it is a problem in our industry.” men [...] It is so balanced Venture Partners is since 1981 and Professor in Entrepreneurial industry more inclusive. Darmon While Bessemer has had several an integral part of the of Management Practice at Management at because no one feels that noted that it was important to grow leadership team at JVP, Harvard Business School, Harvard Business School “wonderful women partners, they… inclusiveness from within. “Very often, overseeing the firms focuses on investments in and a Faculty Research are at a bunch of other places because they are there for any reason it is about mentorship, about growing ongoing operations the software, semiconductors, Fellow at the National they got recruited away from us,” except for merit another generation, giving them room including defining the communications, and Bureau of Economic he said. “But we have more in the to grow, and recruiting in very diverse groups’ strategy as well storage sectors. At Research (NBER). His pipeline.” Darmon asked, “That is places.” Hardymon agreed with as leading the raising of Bessemer, he led early research focuses on a question that you need to explore: Darmon that VC firms need to grow numerous funds that have investments in Parametric financial issues related to Why do they move on?” “Well, because allowed JVP to continue Technology, Staples, startups and high growth diverse talent from within, “and they get attractive offers,” Hardymon building world-class Cascade, Sirocco, Sahara, firms, and their interaction we all have responsibility for that.” technology companies. Celtel, and Endeca, all of with innovation and replied. And Darmon said: “Okay… In addition, he said that Bessemer you just answered the question.” Darmon has over fifteen which led to multi-billion entrepreneurial activity. has “always seen women as being part years of experience in dollar outcomes. Prior to joining Harvard Darmon said JVP has an interesting of the founding group [of portfolio venture capital, strategic Business School, he was He taught and researched situation. “Without ever putting it companies], but we are also seeing planning, capital raising a faculty member at the at Harvard Business School, more and more companies led by and investments in public Stanford Graduate School as an agenda, a statistic, a plan, we where he held the appointment and private companies, of Business. are 50% women partners,” she said. women. We are recruiting more and of MBA Class of 1975 having previously held more women as CEOs, and that is Professor of Management Some of Professor a variety of executive because of the consciousness of the Practice, and served twice Bernstein’s latest research roles with leading Israeli last 10 years. We are hitting the knee as a visiting Professor of includes: Does Venture investment conglomerate, of the curve, at least in some places.” Finance at the London Attract Human Capital Koor Industries, as well School of Economics. and The Creation of Darmon noted that venture as serving as a founding Evolution of Entrepreneurial member of the first Israeli In 2011, Professor capitalists—partners and managing Public Markets partners—need to be rainmakers. corporate VC—Koor CVC. Hardymon received the National Venture Capital Professor Bernstein holds a “And for women to be strong partners, Darmon started her career Association’s Lifetime Ph.D. in Business Economics they must be rainmakers as well,” as a financial analyst with Achievement award. from Harvard University. she said. “The more female-led Claridge Israel LLC, the companies we see, CEOs we see, investment arm of the Now retired from academia, founders we see, we will also see Bronfman family of Canada. he is a co-author of them crossing the lines and becoming Darmon was an Officer Venture Capital, Private partners and leading funds.” in one of the renowned Equity, and the Financing units of the Israeli IDF, and of Entrepreneurship, the holds a BA in Finance and standard textbook for IT and an Executive MBA venture capital. from the Kellogg School of Professor Hardymon holds Management’s international a Ph.D. in Mathematics program in collaboration from Duke University. with Tel Aviv University.

44 COLLER VENTURE REVIEW 45 A New VC Fund Rides the Silver Tsunami

Abby Levy Managing Partner & Co-Founder, PrimeTime Partners

The aging of the world’s population is among the most significant transformations of this century. For Abby Miller Levy, managing partner and co-founder of New York City’s Primetime Partners, this offers an opportunity to build a network of startups that serve the needs of the elderly. Levy’s co-founder at Primetime Partners is Alan Patricof, aged 85, who serves as the fund’s chairman and is a pioneer in the world of venture capital.

he world’s population is that can transform the quality of getting older. This is a living for older adults,” the fund transformation that the was launched in New York City last TUnited Nations describes year at the height of the COVID-19 as “one of the most significant…of the pandemic. In the past six months, 21st century.” The cohort aged more Primetime Partners has invested than 65 is growing faster worldwide in nine portfolio firms that serve than any other group, and this trend seniors or are startups that older will continue for decades. “By 2050, entrepreneurs have launched. The one in six people in the world will be founders are looking for more deals. over age 65 (16%) up from one in 11 “If you have an idea, reach out to us,” in 2019 (9%),” according to data from says Abby Miller Levy, Managing World Population Prospects 2019. Partner and Co-Founder. “We are One expression, among several used in the early days of building a lot to describe this phenomenon, is “the of businesses in this space. No idea silver tsunami.” is too rough to get started.” • For Primetime Partners, this tsunami appears to be less of a destructive force than an energizing opportunity. A $32 million “early-stage venture capital fund that invests in companies...

46 COLLER VENTURE REVIEW 47 The venture community has not Based on what you described as In terms of the segments, these are overlooked this issue; there just has your horizontal approach, how what you would imagine them to not been a robust supply of startups. do you think about its various be. It aligns with what older adults It is a supply issue rather than one segments and components? care about, and plenty of studies of demand. That has limited the have been done to show they care Levy — venture community. The other about three things: their health, There’s data to show how fast the reason is that the venture their financial security, and having population is growing, but because community has become highly businesses aren’t classified by age meaningful experiences. To be sector focused. We have health-tech group—nor do I think they should clear, health is a wide sector funds, fintech funds, consumer- be—there isn’t really a sense of, in ranging from pharma and biotech, tech funds and blockchain funds. aggregate, how big the industry is. which we do not touch. We also At Primetime, the fact that we go There are pockets of specialized do not invest in businesses with horizontal across a consumer industries, like senior living, like regulatory hurdles. We do look at demographic allows us to cross- Medicare Advantage, like some of health IT, aging in place, in-home pollinate several concepts [across How does Primetime Partners view older adults. The reason for that is the government services that serve care, home health aides, et cetera. these sectors]. the opportunities it offers? In this because everything is changing. this population. There are pockets There is a whole industry there. interview with the Coller Venture When life expectancy was 65 and For example, consider direct-to- of data on how fast it is growing, In terms of fintech, this has usually Review, Levy discusses these you retired at 55, there did not need consumer marketing. This is an but that is tempered by how fast been bifurcated into two segments. area from which many founders questions and more. to be all these products, services, the population is growing. The You have financial services: A key have shied away. They do not yet and experiences for a market that population is projected to be 25% question is How do we advise the An edited version of the conversation know how to apply the playbook for people thought was going away. of the global population, up from retirees? To date, that has been appears below: millennials—customer acquisition Either because of mortality or 15% just a couple of decades ago, limited to the very wealthy because cost, life-time value and such because fixed incomes are declining In the U.S., this population controls financial services just have not concepts—to older adults. We have Coller Venture Review — incomes, catering to the elderly was more than 60% of the country’s penetrated mass market older found that we can take one of our What inspired you to focus on not an attractive opportunity. But net worth. adults. Some 50 million Americans investments in a fintech business startups that serve the needs of today, half the population born after do not have a financial plan for that is making great strides in the elderly? Was there an event or 2007 is expected to live to be more retirement. incident that sparked your interest than 100. As a result, we now have a Facebook paid marketing and share in this underserved market, which 30- or 40-year life span that did not that with an e-commerce company Then there are novel financial other VCs tend to overlook? exist before. in our portfolio and vice-versa. products. Other than the 401(k)— the 401(k) was a major innovation — We can call it “retirement” or “ The opportunity in being Levy from the defined benefits and For many people, the catalyst is the golden years,” or whatever horizontally specialized is that There’s data to show how fast the population pension plans—long-term care that they experience either their euphemism you want to use; we can build our own expertise is growing, but because businesses aren’t insurance has been around for a parents’ or grandparents’ failing I personally call it “the Third Act.” in marketing and distribution. classified by age group there isn’t really a long time. Both those markets need health and they get a window into We have a meaningful chunk of There does not need to be a to be disrupted. That is because the health system and how older time in which to provide older brand-new technology that didn’t sense of, in aggregate, how big the industry 50% of Americans are going to run adults are treated. That was not adults with great experiences. exist before to dramatically improve is […] plenty of studies have been done to out of money. The math does not my experience. It was the opposite, That was the impetus. I am an the quality of living for older add up. People are living longer and it had to do with the personal entrepreneur myself, and I started adults. We need to take existing show they care about three things: their but their financial outlook is not growth story for my parents. As to write business plans of possible technologies and better apply health, their financial security, and having changing; in fact, it has gotten I looked at my father, who retired businesses that could fill this space. them, engage older adults, and serve them in different ways. worse. We see a tremendous need early, and my mother as well, As I came up with more and more meaningful experiences That is much more of a marketing for financial product innovation I wondered how they could spend ideas, I realized that my role was and distribution challenge than a as well. their time and what meaningful not best served in creating a technology challenge. experiences they could have. As single business but in building In terms of experiences, this is I began thinking about this, I saw an investment platform to seed Coller Venture Review — where any of the successful startups that other than healthcare and multiple businesses. If you look at the global market of that have been targeting people in housing, there are not a lot of products and services for the senior their 20s and 30s have a corollary. private sector alternatives for cohorts, how fast is it growing? Take fitness, for example.•

48 COLLER VENTURE REVIEW 49 It is a market I know well because Coller Venture Review — authority is a difference in the group. I was surprised that there I used to work at SoulCycle in the As you think about people who go-to-market. were not more older adults fitness area. The fitness needs of are aged more than 60, have you designing for themselves—the way Second, when you talk about older adults are different from those found that this group consumes that every other age cohort seems to being frugal, it’s not about of a 20-year-old. And yet, why aren’t products and services differently design for itself. We have “mommy frugality as much as just the there more robust offerings? We than millennials and younger businesses” being designed by new 50% of Americans reality and the psychology of when are seeing that. In fact, we just consumers? Are they more frugal moms, and college kids creating you’re on a fixed or declining made an investment in a business or are they more willing to spend businesses for college kids, so why income. You have a different are going to run in that space. because they have greater aren’t there more 60- and 70-year- psychology of purchase than when purchasing power? olds saying, “Hey, this is a pain Mental health is another important you know you are getting your next out of money. point in my life. Let me solve this.” area. The way older adults or people Levy — paycheck. You may have people in of different generations view mental That’s a great question. Just to your life who have experienced it. Coller Venture Review — The math does health is different from how the overlay this—and there is some data It has a tremendous impact on Have you found any exceptions? to support this—as we age, we get 20-year-old views it. Taking purchasing decisions. Levy — not add up. People existing products and trying to more heterogeneous. If you think There are ways to get this audience Yes, absolutely. That is a secondary remarket them isn’t as compelling about it, it makes sense, right? You in those situations to become part of our mission, to elevate the are living longer as creating new versions that are get more set in your ways, you are more comfortable with purchasing. role of older adults in the startup designed for older generations. who you are, you are less pressured That can range from installment ecosystem. We have two portfolio to conform. So, when people ask, but their financial The travel segment looks at how plans and parsing things out to companies that are founded by “What are the purchasing habits and where you want to live. If performance-based payments, or all entrepreneurs aged 50-plus. In both of this audience?” I always pause outlook is not you look at Airbnb, more than of those pieces. Another way to deal cases, they took the learnings from and say, “There are so many two-thirds of their hosts are aged with the psychological issues is to their careers and are applying it to segments that to have one point of 45-plus. There is a whole new spread the burden. We are seeing— the older adult market. In general, changing; in fact, view would really, I think, be ageist, way of how people are thinking especially post-COVID—this surge it is something I’m very focused on if anything.” about experiences. of adult children who say, “I want for 2021. it has gotten One of the things that I am to help Mom and Dad, and I don’t I think we are going to continue to Coller Venture Review — continuously reminded of is how know how to do that because how see a big change in the way older Inequality and debt are growing worse. We see heterogeneous this group is. That can I financially support them in adults consume media, commerce, rapidly among the elderly. What said, I think a couple of findings different ways? It is not tax efficient and other aspects of what is often opportunities for innovation does a tremendous are relevant. First, in terms of for me to write them a check. They called the experience economy, this create? What can entrepreneurs purchasing power, this audience don’t want a check from me.” We which I am sure you’ve read or do to enhance the financial well- does a tremendous amount of are going to see more of this kind need for financial written about. But if you think being of older people? research. Whereas when you start of crowd funding. about it, it has not really translated a digital business, targeting a Levy — product innovation to people in their 60s. Coller Venture Review — younger audience, it’s always, “How There are products that are As you continue to research this do you have as little information as around—savings products that market, what has been your most possible, as little friction, to get kick in earlier on. I have been unexpected finding? Has there them to do something?” With older talking a lot about crowd funding. been anything that was truly a adults, it’s “How do we give them So, what is the 529 for older adults, surprise to you? enough information, enough trust for longevity? How do we flip it around? We save, and we have tax authority?” Let them go deep if they Levy — advantages to save for our college want to go deep. Let them pick up I was surprised at how few founders kids’ education. How come we are the phone and call someone if they are aged 60-plus. I think there are not doing that for older adults? want more information so that they people who become consultants can make a more informed decision. or work for themselves or in the gig The biggest cost that older adults That is something we have economy, but there are tremendous are going to face is healthcare. learned from a lot of our portfolio barriers as an older adult to start What most older adults fail to companies. That desire for research, a business, in terms of network, appreciate is that unless you are on for information, and for real access to capital, and just the peer Medicaid and below the poverty •

50 COLLER VENTURE REVIEW 51 represent that perspective. We are now we are up to nine portfolio which is a Zoom enrichment at ground zero on this topic. companies in the past six months platform for older adults. A alone. It has been truly wonderful. bunch of things were accelerated Being a female-run fund, identifying Coller Venture Review — because older adults were online. I am glad you brought up Alan Alan’s ability to leverage his female entrepreneurs is (also) They were shopping online. They Patricof. How did the collaboration experience and his pattern were FaceTiming or Zooming important to me. Female entrepreneurs with him come about? How do you recognition with founders is online, and that also accelerated work together? fantastic, as is his energy for always are attracted to female GPs. There are some of the businesses we were providing new ideas: “Have you Levy — looking at. So, net/net, I would a lot of elements of diversity in addition thought of this; have you thought I have known Alan for a little more say COVID was an accelerant— of that?” Alan also likes to go into to age… gender, ethnic, religious… than four years. I was the founding it was a tragic accelerant, if you the details. If we need to do due president of a business called will—to our launch. We are looking at all those factors Thrive Global with Arianna diligence on a company, he’ll make Coller Venture Review — Huffington. Alan Patricof was an the customer calls. Alan likes to You mentioned nine portfolio investor in Thrive, so we met in that play the role from analyst to companies so far. What did you context. Just over a year ago, I was managing partner; he spans the look for before choosing to invest having brunch with one of his sons, range. I have thoroughly enjoyed in these ventures? who I went to business school with, working with him. We have a few and I told him that I was starting a other team members that have Levy — venture fund focused on older subsequently joined the team, We look for what every other adults. He put down his fork and and I would say—and he would venture capital firm looks for— say—he is having the most fun he a compelling product with line, your home health costs aren’t Coller Venture Review — has ever had in his career, and this evangelical users and a great covered, and there’s so much When older people move from is where he wants to have founding team. There is nothing out-of-pocket expenditure. So how careers into retirement, in theory tremendous impact. different about our method of can we create new supplemental they have more time than younger Coller Venture Review — investing. As a first-time fund, insurance products that prepare generations. What opportunities You mentioned the pandemic. What almost creating a new sector, we us for the burden that is to come? for leisure products and services impact did that have on your plan plan to be the biggest and most does this create for entrepreneurs? One other aspect involves real beyond the timing factor that you significant. For us, the strategy Have you found entrepreneurs who estate – and we just invested in a mentioned? Did you have to revise was to make sure across our three are willing to help create second business that relates to this. For your strategy in any way? areas of health-tech, fintech, and and third careers for retirees? most adults, their largest piece of consumer-tech, that we were Levy — their net worth is their home. One Levy — placing some bets on platforms that That’s a great question. I say almost of the companies we have invested This is something that I feel have the potential over the next sheepishly that the pandemic sped in, Fraction, has a new type of passionate about. Alan Patricof decade to have tremendous impact. up our ability to go to market and mortgage product—a fraction and I started [Primetime Partners] accelerated everything we were Sometimes that would mean appreciation mortgage—because when I was 45, and he was 85. doing. When we went to investors smaller check sizes than we would we must find more ways to unlock People ask all the time, “What’s it saying we were raising this fund, it have liked. Sometimes it might have the value that these older adults like working with Alan?” And I say, was while a light was being shone been at around a little later stage have created. “We talk every morning at 7:30, on the issues facing older adults in than we would have liked. But the after he has done his 10-mile bike The last area that I hope helps with terms of what is happening in key point was that we need to start ride. We work a full day.” I think the financial issue is employment. senior living and nursing homes. to build this ecosystem of founders we need to recognize that every How do we find alternative income How do we take care? How do you that we can work with across the person has strengths and sources for older adults? When I said, “That’s what my dad wants to age in place? All those issues were portfolio. That was really what the weaknesses, and it is irrespective think about some of the biggest do.” It was quite a timely, I would front and center in a way that I strategy was, honestly, to get going. of age. The sooner we do that, we opportunities, most of these questions say, fateful moment. Alan and I had would just say everyone nodded can start to recognize, “Here’s what Coller Venture Review — we have raised have been the focus lunch the following week, and that their heads. Let us put it this way: this individual has to contribute,” How do you set yourself apart from of the nonprofit and government was that. There’s not a single person that we rather than judging them by other funds that are also focused on sector. This is because to date, they spoke to who did not get it. I think characteristics such as age. Within a few months we formalized, the senior market? have been focused on and, frankly, we secured a partnership with that was certainly helpful. bearing the financial burden of our I think we have not even begun Levy — Welltower Senior Living Real And then, some of our businesses aging population. Now I am excited to start that process as a society. I We do that in three ways. One is Estate Investment Trust, and we were accelerated by COVID. We are to see more private-sector founders hope through our work to continue my background as a marketer and were about to launch in March invested in Tembo Health, which saying, “How can I help the to show role models who are ageless as an entrepreneur myself. I think when COVID hit. We delayed nonprofit and government sectors? and contributing. We also hope is a specialty telemedicine business this is mainly about distribution fundraising for just a couple of How can we work together?” we can hire folks on our team that focused on senior living facilities. and marketing. I am very hands-on months but started investing, and We are invested in GetSetUp, with the founders and the teams •

52 COLLER VENTURE REVIEW 53 on product messaging, marketing, portion will be started by older with these issues is to demonstrate and they know we can help them and growth partnerships. adult founders? As many as we can through role models and action and grow. The third element is more find is the answer. Right now, we results that this is a market worth personal. It is to build a firm that The second piece is building this have two out of nine. We do not designing for. I love that we are stands by the values of our next gen group of founders with have any quotas or thresholds, but talking with government agencies. founders. Starting a firm is an whom we can work across the that’s what we’ve identified as being My point of view is, “Let’s prove it entrepreneurial task in and of itself. portfolio with synergies, such as great investments. It is something We are a startup. When I look at our referrals and sharing learnings. through showing success.” we track—the diversity of all of our success, it is not just the success of Coller Venture Review — The third piece is thought founders. Being a female-run fund, our portfolio and our brand but also As you explore these opportunities, leadership. We did some press identifying female entrepreneurs of our team. That means working what do you think are the biggest when we launched. You will see a is important to me. Female with great people and helping them risks in targeting the senior market? lot of other PR-worthy activities entrepreneurs are attracted to fulfill their goals, whether you are a What are you doing to mitigate from Primetime in 2021. We want female GPs. There are a lot of 25-year-old associate or an 85-year- those risks? to be a beacon in the space, and so I elements of diversity in addition old managing partner, that we are view our job as two-thirds investing to age... gender, ethnic, religious, Levy — all having enriching and rewarding and one-third leadership. I think you can go on and on. We are It is so specific to each business. careers. Those would be three About that is a little different. The founders looking at all those factors. We are still at the point when elements of success for the fund. we speak with recognize that we people talk about aspirational All the VCs I Abby Levy, Managing Partner & Co- Coller Venture Review — My message to anyone who is are going to be bold in trying to marketing. People still think about Founder at PrimeTime Partners has As you were speaking with have spoken reading this is, if you have an idea, build this category. There have the silver-haired man and woman spent her career helping businesses and investors about participating in reach out to us. We are in the early consumer brands grow as an operator, been other funds there that are walking down the beach, swinging to have had your fund, how difficult was it days of building a lot of businesses entrepreneur and advisor, most notably wonderful partners to us but have their arms together. This notion to convince them not to fall prey a bullet point in this space. No idea is too rough in the wellness sector. Prior to Primetime been much more focused just on of the realism of aging is a risk the investing activities. to ageism and be biased in favor to get started. Partners, Abby Levy was an executive at of younger entrepreneurs and because it is not as beautiful as it on their list SoulCycle, where she oversaw business Coller Venture Review — founders? Why does this bias has been depicted. There is a for a few years development and revenue growth outside As you think about your investment against ageism persist? marketing challenge around the consumer studio business, with an philosophy, would you say you are What can be done to combat it? aspirational versus real. I do not called “the emphasis on building new digital products more motivated by social impact or usually think about it in terms of silver tsunami,” as the Senior Vice President of Strategy financial potential or both? Levy — risks. Part of the reason is that there & Growth. To answer your first question, is so little built, and it is such a big or “longevity,” Levy — Abby Levy has also been a founder, our investor group believes whole- space. For startups, just pick a piece teaming with Arianna Huffington to We are a for-profit fund. We work or “the silver heartedly in our mission. If you of the puzzle that you want to solve. launch Thrive Global, a behavior change on behalf of our investors, and look at the demographic trends, Solve it well, and you will have a big market” technology company focused on employee financial impact is our priority. it is a no-brainer. All the VCs I business. The only risks I would see productivity and wellness. Abby Levy However, if our portfolio companies have spoken to have had a bullet are people stumbling in terms of served as President of Thrive Global do not have social impact, they are point on their list for a few years and remains on the Thrive Board. not doing their jobs right. You will how to connect with this audience, called “the silver tsunami,” or Abby Levy began her career at McKinsey not see us investing in businesses but not anything structural, not “longevity,” or “the silver market.” & Company then led product development that just serve an exceedingly small, anything competitive. It is a blank We did not come up with this canvas for startups. at OXO International. She is a graduate affluent portion of our society. brand-new idea. We are just of Princeton University and Harvard Almost all our businesses serve executing against it, and our Coller Venture Review — Business School. the mass market of older adults. investors completely believe in it. Where do you expect Primetime That is both because that’s where Partners to be in the next three In terms of ageism, that is one of the financial opportunity is, and to five years? How will you define the -isms that is not really talked secondly, it’s where I think our your success? values are most aligned with our about yet. There are several reasons founders. There is no reason you for that. It just has not had visibility Levy — can’t do both, but we lead with what in the same way. It is interesting, Our success comes when we have is a good financial opportunity. because we just went through a from our first fund a cohort of presidential election where both companies that are big businesses, Coller Venture Review — the candidates were aged more household names that are viewed How much of your fund will be than 75. There has not been really a as founders who transformed this invested in ventures launched by tremendous visibility to this issue. sector. That is one element of older entrepreneurs, and how much Why? There are all sorts of reasons, success. The second is that any will be services for the elderly? everything from our own fear of our founder in the space who wants Levy — mortality to the concerns around to start a business that serves older All our businesses need to serve the entitlement economy. My adults comes through our doors older adults. Within that, what perspective on how to get started because they want to work with us,

54 COLLER VENTURE REVIEW 55 Overview

iv ur Trends in Venture section addresses 59 change and challenges in new venture creation. Emphasizing a Data-Driven This issue considers opportunities for new entrants, both among individual entrepreneurs View in New Ventures O and within emerging sectors. Professor Ethan Mollick Ethan Mollick addresses the so-called “unicorn’s myth,” Associate Professor of Management, and the apparently false assumptions uncovered in his Ralph J. Roberts Distinguished Faculty Scholar, research that seem to persist among entrepreneurs. Wharton School of the Most significantly, Mollick makes a strong case for a University of Pennsylvania more inclusive path and towards the democratization Trends in Venture of entrepreneurship. From an industry perspective, he is joined in this section by Xavier Vives of IESE, who addresses the opportunity for new venture, specifically 67 in the form of FinTech innovation. What are the FinTech and the Together, these contributors combine theory and practice Future of Banking to help us consider new opportunities at an individual and sector level. Looking forward, future discussions in the Professor Xavier Vives Trends in Venture section will continue to compare and Professor of Economics and Finance, contrast thematic change, including innovations and their Entrepreneurship Myths IESE Business School practical implications at a macro and micro level. that Deter Entrepreneurs?

56 COLLER VENTURE REVIEW 57 Shifting the Entrepreneurial Paradigm to a Data-Driven View

Professor Ethan Mollick Associate Professor of Management, Wharton School of the University of Pennsylvania Author of The Unicorn’s Shadow: Combating the Dangerous Myths that Hold Back Startups, Founders, and Investors

n this interview with the Coller Coller Venture Review — Venture Review, Wharton What inspired you to write a book Associate Professor Ethan Mollick about myths of entrepreneurship? Iemphasizes the importance of Did your experience as an methodical testing and measured entrepreneur shape your thinking adaptation among startups in a post- before you entered academia? COVID-19 world. Mollick discusses Mollick — the most enduring myths about It did—but my ideas were shaped entrepreneurship and how recent more by teaching thousands of research shows that these are false. students about entrepreneurship. He also speaks about how founders It is such a powerful concept. People get ideas for new ventures, how to want to become founders. They make better pitches for capital, and read a book about Steve Jobs or what the world can learn from Israeli Mark Zuckerberg or Peter Thiel, startups. Most importantly, he offers and they try to emulate these advice on which skills founders models. Looking at such examples should build in order to help their is fine, but the issue is that in the last ventures scale up and succeed. 10 years, there has been a boom in entrepreneurial research that uses real data to get at deep questions about how you start a company and what you should do as a founder. Data often contradicts some of the received wisdom. That is what Myths about who should encouraged me to write this book. Many people become discouraged become an entrepreneur from entrepreneurship because are dangerous because they follow the wrong advice. they discourage people Coller Venture Review — What do you consider to be the who otherwise might start wrong advice? What are the most companies from doing so dangerous myths and monomyths about entrepreneurship? Where do they come from and why do they persist? •

58 COLLER VENTURE REVIEW 59 Mollick — shows that older founders are more its meteoric rise? Certainly, some I separate them into myths about successful. Another myth is that of it might have been because of whether you should become an you must have a certain personality This does not mean that every the clever strategy of launching in entrepreneur, and those about how type to be a good founder; we have colleges first and creating demand. to be successful as an entrepreneur. lots of evidence that that is not company should be founded by solo But it also was at the right time Myths about who should become true—there is no entrepreneurial entrepreneurs, but there is little at the right place. People were an entrepreneur are more dangerous personality type. Such myths are connected to the internet and because they discourage people dangerous because they discourage support for the view that you should found it interesting. What can you who otherwise might start companies people from even trying. always found companies in teams replicate from that? If you just take from doing so. That is a problem lessons from these unicorns, you Another big myth is that once a because we already know a lot of might end up emulating things company is launched, you should biases exist. Founders of venture- that may or may not have been just start doing stuff, just go out backed, high-growth companies responsible for their success. and start selling. But there is a are disproportionately male, white, process that entrepreneurs should Coller Venture Review — and highly educated. It would be go through when they want to run You write about the importance nice to see more people from other a company, a process of disciplined of using data to bust myths. Could backgrounds launch companies. 20%–50% experimentation. It is not about you offer more examples of how new scholarship is overturning The boost to your chance of success The myth that you must launch a just starting to do things. You by writing a business plan even though company as a young techie does need a disciplined approach. conventional wisdom? they are out of style not match up with the data, which Mollick — Sure, the book is full of data like Coller Venture Review — with you or give them equity stakes that. For example, you get about One of the issues you address without them being co-founders. a 20% to 50% boost to your chance in your book is the need for co- of success by writing a business Coller Venture Review — founders, which venture capitalists plan even though they are out of What kind of shadow do unicorns style. That is not because business seem to almost insist on. What cast over the entrepreneurial plans are useful as your plan, but does the evidence show? ecosystem, especially in places they help you as a team surface like Silicon Valley? What myths Mollick — issues and coordinate on the future. do unicorns breed, and what are I have a research paper with Jason That is a useful process. I talk Greenberg at New York University the implications? about data on the age of founders “Sole Survivors: Solo Ventures Mollick — and the psychology of founders, Versus Founding Teams” where Unicorns—or private companies and where all that comes from. we show that solo founders often worth more than $1 billion—are the One of my favorite examples is the outperform groups of founders. lottery ticket winners. Those are the data showing overwhelmingly, that This does not mean that every touring rock bands or the Hollywood if you want to be more creative, company should be founded by actors, to use analogies from other both sleep and coffee help. That solo entrepreneurs, but there is fields. That is why people go into is useful information. Everything little support for the view that you acting, or why they start playing in the book is data informed. should always found companies in the guitar, or why they become Coller Venture Review — teams. A lot of evidence suggests entrepreneurs… they believe they Coming up with ideas for a startup that founding a company as a team are going to be unicorns. Part of is often the first challenge for an can create all kinds of tensions. If the issue with unicorns is that entrepreneur. Where do founders you have someone with whom you becoming one requires a lot of come up with startup ideas? are launching a company, it is often luck and good timing. When you Mollick — difficult to resolve those tensions. look back at unicorns that were In the book I discuss a few different A study from about 20 years ago successful and you try and emulate ways, because there is not one them, not only are you at risk of surveyed venture capitalists. It magical method. Often a good way emulating bad behavior—what showed that when a company to start is by scratching your own we call superstitious learning— fell apart, 65% of the reasons had itch to think about an effectual but you are also not necessarily to do with conflicts within the approach to entrepreneurship. emulating the core of what made senior management team. That is What do you know, who do you that unicorn successful. a common way for companies to know, what resources do you have fall apart. People are too willing to For example, think about Facebook. at hand, and what can you build emphasize having co-founders. In It was not the first social network. from that? You are probably an fact, you can hire people to work What lessons can you learn from expert in something, whether it •

60 COLLER VENTURE REVIEW 61 the changes to geography. For should make you successful at most venture capital funding, the raising capital. You want to show average distance for companies the experiments that you have in which VCs invest is only 80 been conducting are successful, miles, and 40% of VC investments and what they have taught you, are within two miles of their and how you can continue to headquarters—so investment has learn and grow in the future. been highly centralized. People Coller Venture Review — thought, maybe there will be a drop Many new ventures flounder in capital investment, but maybe because, though they start strong, The first thing you need to do is know then we will be using Zoom to make they fail to scale. What are some of investments and we may not need how to tell a good story. [...] A lot of the biggest challenges that startups to go to Silicon Valley anymore. face in building their operations founders overemphasize showmanship Both those things have turned out over time and across regions? How not to be true. So far, the change in over preparation should they deal with them? capital investment has been small, and the capital is still going to the Mollick — same old places it always went to. The most important thing is organizing yourself early. Coller Venture Review — Companies tend to continue along Interesting. The more things change… their initial path. Founders who Mollick — do a good job setting up their Exactly. organizational structure and caring about it from the beginning Coller Venture Review — generally do better. A paper that Pitching to angels, VCs and other just came out shows that you potential investors is among the cannot make up later for missed most important skills for any organizational opportunities at entrepreneur. What are some the beginning. Figuring out how of the most common mistakes is a hobby or a particular task at Then you have massive dilutive you want to grow your company, that you have seen founders your job. So where are the gaps? funding, which includes things like how it will look, what the org make in this regard? How venture capital or angel investment. chart will be early on is helpful. In Another way I like to give people should they correct them? There is a lot in the book about a hint is when you’re at a job and addition, hiring methodically, using how you can sell and pitch to these you see Microsoft Excel being used Mollick — proven techniques like behavioral for anything other than adding groups of people. But the general The first thing you need to do is questions, can be extremely helpful. up numbers, it’s almost always an goal is, only raise as much as you know how to tell a good story. On the customer side you need opportunity for a piece of software. need and only when you need it, That often means sticking to a to make sure of the product-market This is because Excel is the which is easier said than done. standardized method by which you fit. Make sure you understand why programming language available But it’s not a trophy; it’s a means to tell these stories, a set of slides you your customers are buying your to the masses, so it’s a great place an end. Your venture is an engine— use. I would recommend sticking product before you start spending to start to find a business idea. and capital is fuel for your engine. to that method, but also making a lot of money on marketing. Your job as an entrepreneur is to sure you can tell that story in a Coller Venture Review — Coller Venture Review — build the engine and then get the compelling way. A lot of founders As founders think about the human Capital is the fuel that drives right fuel for it at the right time. overemphasize showmanship over side of scaling, you said that startups. What does the evidence preparation. The evidence shows Coller Venture Review — many ventures fall apart because show about how entrepreneurs that for professional investors, Has the ability of founders to of conflicts in the management should raise capital to keep showmanship, like passion, does raise capital dried up because team. As a company scales, how their ventures growing? little, while preparedness and of the pandemic? If so, how should it integrate the right kind Mollick — organization and clarity does are entrepreneurs dealing of management expertise so There are two flavors of capital— a lot. For amateur investors, with that situation? that this does not happen? dilutive and non-dilutive. Non- showmanship can often work dilutive capital tends to be the Mollick — better, but for professionals— Mollick — founder’s own money, or resources The latest numbers just came out that is the kind of investment It is important for founders to raised through crowd funding or a little while ago. People were that founders are looking for— be aware of their own skill sets grants. Those are all nice forms of predicting that there would be a you want to show methodical and to bring in people who have capital because you do not have to big drop in the amount of capital testing. The things that make you complementary skills early on and give up any equity of your company. being invested and worrying about successful as an entrepreneur have a clear span of control for •

62 COLLER VENTURE REVIEW 63 We know that startups that improvise tend to survive. In contrast, startups that have too many resources fail at a higher

those people. Knowing what they You want to be unique, you want to the classroom and use them to help rate than startups with less resources. are responsible for helps prevent tell your own story, but I think you transform education. One of our first You don’t want to have no resources, but conflict and helps you scale up well. should be aware of what the data products is called Saturn Parable, Founders also need to constantly let says so that you can make the right which takes place on a fake mission you don’t want to have too many. In the go of things as they grow, because choices about when to ignore it. to Saturn in 2087. A lot of the stuff middle, there’s improvisation—and figuring they cannot have their fingers in about it is realistic because we want Coller Venture Review — everything the way they did when the science to be real. But the idea out what to do with what you have What is the Saturn Parable, and the company first started. Having a is that, if you put people in a fake how does it tie into the themes About structure where you are training and setting, they often learn better than of leadership and teamwork handing things off to people from if it is a realistic setting—especially that you discuss in the book? the beginning is extremely helpful. about leadership or teamwork. We Professor Ethan Mollick is an Associate Professor of Management and Ralph J. Mollick — have you play this simulation, and Coller Venture Review — Roberts Distinguished Faculty Scholar In addition to this book, I wrote crazy things happen that stress What are some of the most at the Wharton School of the University you in all sorts of interesting ways, important lessons you have another one a while ago on video of Pennsylvania, where he teaches games in business, a subject that I and they force you to make errors learned about evidence- of need to improvise. We know A study in Italy found that innovation and entrepreneurship. He is have been interested in for the last 15 with your team that are the kind of based entrepreneurship that startups that improvise tend companies that learned how to an award-winning author of numerous years. It is about how you use games errors that happen in real life but in publications, recently The Unicorn’s based on your research and to survive. In contrast, startups run experiments in a scientific way and simulations to teach effectively. an accelerated way. It’s a really cool Shadow: Combating the Dangerous Myths that of other colleagues? that have too many resources fail at for their startup had twice as high I have been teaching using an discussion. Think of it as an escape that Hold Back Startups, Founders, and a higher rate than startups with less revenue a year later as those that Mollick — entrepreneurship simulation in my room meets leadership training Investors. His work on crowdfunding is resources. You don’t want to have did not take a scientific approach. The most interesting paper I recall classes for the last four or five years. experience meets a classroom. the most cited article in management no resources, but you don’t want Those are two clear skills that is one that showed that most stories Students run a realistic fake startup published in the last five years. Coller Venture Review — to have too many. In the middle, founders could focus on to be more that people tell about why their company over the course of several As the Academic Director and Co-Founder Israel is often called a startup there’s improvisation—and figuring successful. What I urge all founders startups failed end up not actually weeks, and they must manage all of Wharton Interactive, Professor Mollick nation because it has more startups out what to do with what you have. to do is realize that you must design telling you the root cause of why the processes from legal issues to works to transform entrepreneurship they failed. They are to help the per capita than any other country. That kind of outward focus of your startup so that you can always customer issues. I bring in a mix of education using games and simulations. emotional process of grieving for a What lessons can entrepreneurs having a small market that you need keep learning. You have to learn actors and real people to play roles. He has long held interest in utilizing games company from the founder. It’s hard in other parts of the world to quickly overcome is one of the about yourself as a founder. in education, and he co-authored a book I have professional dungeon masters to learn stuff from just observing. learn from Israeli startups? reasons why Israeli companies have on the intersection between video games who play various characters. been so hungry and successful. If you want to learn things, there Mollick — and business that was named one of the American Library Association’s top 10 is a strong mismatch between the I have been working with a new Israel is either third or fourth in the Coller Venture Review — business books of the year. He has built popular view and what the data group at Wharton that I founded expanded European, Middle East One of the most important numerous educational games, which are suggests. I think that people do not along with my colleague Sarah area in terms of fundraising. One messages from your book is that used by tens of thousands of students want to necessarily know, and I do Toms called Wharton Interactive. of the most powerful lessons of Israel entrepreneurship is not in-born, around the world. not think the data is the only answer. The goal of that group is to take is the importance of ecosystems. but a skill that can be taught and games we have been running in They are not easy to start—you learned. What advice would you Professor Mollick holds a Ph.D. and MBA from MIT’s Sloan School of Management cannot just start an ecosystem. give founders about the most and a B.A. from Harvard University, magna You need successful founders, important skills they should learn cum laude. sources of capital and government to succeed, even if their firms are The things that make you successful as an investment. Then it becomes not unicorns? self-sustaining in a useful way. entrepreneur should make you successful at Mollick — Starting with community, with We know that the two things raising capital. You want to show the experiments a group of people and creating you can learn that will make you that you have been conducting are successful, a virtuous cycle, is an important more successful as a founder— lesson from Israel. The other thing based on actual controlled and what they have taught you, and how you can that Israeli startups teach us is experiments—are pitching and continue to learn and grow in the future about improvisation and bricolage. experimenting. You can learn In a place where there were not how to do pitching better, and it necessarily a lot of resources or has an effect. You can also learn capital to get started, there was lots how to run experiments better.

64 COLLER VENTURE REVIEW 65 FinTech and the Future of Banking

Professor Xavier Vives Professor of Economics and Finance, IESE Business School

he current rapid pace of Historically, banks have controlled change in the banking sector digital forms of money and payments appears to be accelerating, through regulatory protection of Twith digitalization deposits, exclusive access to the increasing competition, including central bank settlement system, in the adoption of different forms of and partnerships with credit card digital currencies. Combined with companies. However, the challenge low interest rates and the resulting today comes from a variety of digital pressure on bank profitability, the assets that do not sit on the balance potential short-term implications sheet of banks such as electronic The potential short- for entrepreneurial startups in the wallets, stablecoins, and balances term implications for financial system has never been larger. with a telecom provider. Control of these digital assets are among Among related developments, we new entrepreneurial the competitive advantages of new see the massive application of digital entrants. Specifically, payment startups in the technologies and the emergence of technology—including connections new competitors. These have allowed financial system has to the growing digital life of consumers for many new products and services and business alike—has been among never been larger and helped improve the efficiency of the new entrants’ keys to success. incumbent banks. At the same time, they have also favoured the entry of Technology-based disruption has of new firms, increasing competition course occurred outside of just the with traditional bank business payment paradigm. For example, the models. Startups face uncertain entry of new players offering credit has interaction effects, the result and provided another foothold, affecting impact of which depend on the market banks through downward pressures on structure that eventually prevails. fees and prices and more compressed margins. This is particularly true when a country’s general development •

66 67 Both banks and customers have realized that they can work and operate remotely in a safe and efficient way. The response to increased digitalization will require more investments in IT and will leave obsolete the overextended branch network of many banks, particularly in Europe

is higher and its banking system is platform provider. In such a future, 1 E. Carletti, S. Claessens, A. Fatás, About less competitive, but less so when the the degree of rivalry in the market X. Vives, The Bank Business Model in the Post-COVID-19 World, Centre country’s regulation is stricter. Across will depend on the cost of switching for Economic Policy Research, 2020, Professor Xavier Vives is Professor of the board and by contrast, non-bank from one ecosystem to another. The https://media.iese.edu/research/pdfs/ ST-0549-E.pdf Economics and Finance, Abertis Chair entry has not yet been meaningful larger the costs, the less competitive of Regulation, Competition and Public in demand deposit-taking activities, the market will be. In this world, Policy, and Academic Director of the Public- possibly due to concerns about dominant platforms could include Private Research Center at IESE Business regulatory burdens. Change has been current digital giants plus some School. His award-winning research focuses uneven, but steadfast in its growth. platform-transformed incumbents. on industrial organization and regulation, economics of information, dynamic rivalry, Regulators have adapted to the A main conclusion of the recently innovation and competition, banking released report The bank business digitalization disruption by balancing crisis and regulation, and information model in the post COVID-19 World1 is their fostering of competition to allow and financial markets. Most recently he that pre-COVID-19 tendencies will the benefits of innovation while still co-authored The Bank Business Model be accelerated, as subdued economic protecting financial stability. They in the Post-COVID-19 World. must ensure a level playing field and growth and low interest rates persist Previously, Professor Vives served as coordinate prudential regulation and and digitalization in its many forms a member of the Advisory Board for continues to advance. The response competition policy with data policies. Economic Recovery of the Government to increased digitalization will This will require navigating complex of Catalonia and Special Advisor to the require more investments in IT and trade-offs between the stability and Vice President of the European Commission integrity of the system, efficiency will tax the overextended branch and Commissioner for Competition. He and competitiveness, and privacy. served as President of the Spanish Economic network of many banks, particularly Association, Duisenberg Fellow of the in Europe. The result is that the sector European Central Bank, and President will need a deep restructuring. of EARIE. The swift shift towards a more digital Professor Vives has been an advisor and world in response to COVID-19 makes consultant on competition, regulation, and clear that the speed of change may take corporate governance issues for numerous everyone by surprise. Interestingly, global organizations including the World large technology companies have many Bank, the Inter-American Development of the ingredients to get ahead in the Bank, the European Commission, and the post-COVID-19 world. They are digital Federal Reserve Bank of New York. natives, have the technology, customer In this world, dominant platforms could Professor Vives holds a Ph.D. in Economics base and brand recognition, as well as from UC Berkeley. vast amounts of data and deep pockets. include current digital giants plus some A possible outcome is that a few platform-transformed incumbents dominant platforms will control the access to a fragmented customer base, each of which inhabits different ecosystems. In this case, financial service providers will compete to supply the platforms, and customers will demand services from their

68 COLLER VENTURE REVIEW 69 Overview

v ur Industry Analysis section draws on one 72 of the most essential tools for analyzing a Leading Transformative commercial ecosystem. The tool, conceptualized and then refined over decades, facilitates an Change in Digital Health – O understanding not just of what is, but of future transformation Lessons from Practice within an industry and, often, across industries. Dr. Lilach Weisz In this issue, we appropriately enough address digital health, Head of Innovation and Tech Transfer, and the opportunity to improve healthcare accessibility Tel-Aviv Sourasky Medical Center and efficiency by bridging the practical chasm between technologists, entrepreneurs, and healthcare institutions. Tamar Many Industry Analysis Co-founder, MindState We are joined by Lilach Weisz, Head of Innovation and Senior Design Lecturer, Tech Transfer at the Tel-Aviv Sourasky Medical Center, Shenkar College of Engineering, and Tamar Many, Co-founder of MindState, an ideation Design and Art lab focused on digital health and a Senior Design Lecturer at Shenkar College of Engineering, Design and Art. A Shot in the Arm for The authors underline the ways in which scaling new starts in digital health require collaboration across fields to instantiate theory into practice. Looking forward, the Industry Analysis section in future Digital Health Innovation issues will similarly be written by industry leaders addressing technology-driven innovation.

70 COLLER VENTURE REVIEW 71 Leading Transformative Change in Digital Health – Lessons from Practice

Dr. Lilach Weisz Tamar Many Head of Innovation and Tech Transfer, Co-founder, MindState Tel-Aviv Sourasky Medical Center Senior Design Lecturer, Shenkar College of Engineering, Design and Art

As leaders of healthcare While venture investment has grown innovation, our close work to embrace digital health, the field still with seed stage digital health startups appears to lack consistent and clearly has illuminated insights regarding defined paths to support new venture digital health innovation. From our success. A key underlying question is experience, startups that have partnered how to improve the path to innovation with healthcare organizations, as where there is the potential for early as the research stage, have been transformative change. When we probe able to innovate, adapt, and execute further, we find that the gap between nimbly. In this article we discuss the healthcare centers and the technologies basic tenants of such partnerships, meant to support them remains quite including points which can help broaden significant—and that it is within accessibility and deepen efficiency.” this gap that many digital healthcare The field of “digital health” is defined startups seeking to innovate lie. as the combination of healthcare A few case studies drawn from our and technological developments work can underscore the importance that deliver the next generation of crossing the chasm between practice of digitized healthcare solutions. among digital health startups, and their While the field may be only a decade brethren working on site in hospitals old, it is rising in significance. This and related healthcare settings. is partly due to the impact of the recent COVID-19 pandemic. More GistMD: GistMD’s mission is to fundamentally, the rise of digital create a smart platform that provides health reflects the need to improve patients with personalized information healthcare efficiency and accessibility. about their medical journey, including Put simply, the field will likely play explanatory videos about diagnoses a significant role in bringing about and procedures. The platform engages the next era of healthcare delivery. patients in their medical processes, •

72 COLLER VENTURE REVIEW 73 leading to reduced patient anxiety, of healthcare, tapping into the greater patient satisfaction and time “innovation ecosystem” is one way to saving for medical staff. To date, do so. Hence the timelines and specific GistMD has raised $2.2 million. contours of digital health-healthcare They have also successfully deployed partnerships are worth exploring The field still appears their platform at a world-leading bilaterally. Below we therefore suggest research hospital and are in the elements of best practice which to lack consistent advanced stages of deployment we believe can help build a solid at other leading hospitals, who foundation for potential partnerships, and clearly defined will become the firm’s first paying propelling startups and healthcare customers. The transformation organizations alike towards innovation. from initial bedside observation paths to support to a funded digital health startup, would have been unlikely absent 1. Finding the Optimal Partner new venture success. their early, hospital-centric trials. Once a seed stage digital health startup decides to partner with a healthcare Enroute: Enroute addresses intra- organization, it is crucial to find those 3. Not Just Doctors A key underlying hospital transportation systems capable of facilitating its growth. This characterized by long waits for patients Partnering with healthcare question is how and staff. The venture’s intelligent specifically means the organization organizations, on the one hand, is a algorithms dispatch transporters on- should be willing and able to provide matter of finding physicians who deeply to improve the demand to deliver patients and medical in-kind resources (e.g. time, manpower) understand the startup’s ecosystem equipment, and transparently provide or actual funding (e.g. via VC arms and unmet needs they are tackling. path to innovation this data to medical staff. Their hospital for later stage funding). An in-house But this partnership is by no means pilot is expected to demonstrate a startup collaboration unit dedicated to just about physicians—it’s also about significant delta in accessibility and supporting practice-based partnerships understanding the complex group of where there is efficiency, ahead of funding and can provide additional significant relevant healthcare stakeholders and geographic expansion. Absent the benefit. Startups will have to interface finding ways to optimally interact with the potential for opportunity to trial and transform with multiple stakeholders across the each. Depending on the project, the the venture’s underlying algorithms organization, and an internal team can network can include psychologists, transformative change based on observation in practice, this facilitate this. It is also helpful for the social workers, cleaning/cooking staff, innovation would likely have stalled. healthcare organization to overlap as well as IT and financial departments, with the general aims and expertise regulatory bodies and more. A key While the above two startups may not of the startup. message here is that startups will be exactly what one thinks of when have to interface with multiple medical considering innovation, both teams across varying divisions, capture and reflect the challenges that 2. Building a Long-Term including diverse stakeholders healthcare organizations face. They Strategic Partnership and numerous internal systems. are precisely the type of innovations A key goal of a partnership between that are ripe to be incorporated technology startups and healthcare into healthcare organizations— centers should be to build the 4. Champion and, indeed, both these startups foundation for long-term collaboration. The notion of a “champion”— partnered with hospitals at early Designing and implementing a particularly one from within stages, with positive results that formalized agreement can help the healthcare organization—is may not have been predicted. healthcare centers function almost crucial to digital health startups. It may also seem risky for healthcare like a startup incubator, giving the These champions will lead from organizations to partner with seed startup room to learn, grow and within the healthcare organization, startups such as GistMD and Enroute. explore. If successful, the organization becoming actual startup partners But as many top organizations are even has the potential to become the who work diligently throughout constantly searching for what can startup’s first paying customer. the startup journey. keep them on the cutting edge

74 COLLER VENTURE REVIEW 75 Practically, this means helping build roots within the organization by creating an in-house team which spans While partnerships are across the organization – clinical, administrative, and technological. not quick and easy nor a The expertise of this champion can guarantee of success, they range depending on the startup’s needs, whether a medical practitioner, tend to challenge the rigidity operational manager or IT lead. of the silos which separate They should also be an opinion leader within their field, envisioning technologists from their where the startup exists within the eventual contexts, thus ecosystem, and an early adopter, not resistant to industry change. improving outcome measured 6. Process-Oriented, by the impact and scalability About 5. Digital Health Data Human-Centric, of any given innovation Digital health startups depend on rich Multidisciplinary Teams Dr. Lilach Weisz is Head of Innovation data throughout their journey. For this Seed startups need to move quickly and Tech Transfer at the Tel-Aviv Sourasky reason alone, there is no better place to develop a rough idea of how to Medical Center. Dr. Weisz holds a Ph.D. in Cancer Genetics from the to find healthcare data than within the solve unmet needs. But too often they Weizmann Institute of Science, an MBA healthcare organization. Healthcare spend months working in isolation on in BioTechnology from Colman Business data runs the gamut, from patient a prototype which doesn’t meet real School, and an MS.c. in Biochemistry health to demographic information world needs. For this reason, digital from Tel Aviv University. to operations. For digital health health startups should not focus on startups, filtering through this data to solving specific outcomes, rather on Tamar Many is Co-founder of MindState, excavate insights is key. Partnering the process of partnering early and an ideation lab focused on digital health, early also means being able to use data learning from healthcare stakeholders, and a senior design lecturer at Shenkar to help validate basic assumptions and then developing quick prototypes College of Engineering, Design and Art, surrounding unmet needs. Finally, 1 1 towards iterative solutions . The Path Forward https://medicalfuturist.com/5-things- and a lecturer at the Coller School of new data may be collected along the we-learnt-about-investments-in-digital- health-new-e-book/ Management at Tel Aviv University. startup’s journey—both quantitative To ensure that startups don’t over- In the coming decade, digital health and qualitative—which is a value- engineer solutions that real humans innovation will continue to grow. is a Research Assistant. add for the organization. don’t need or want, much care should Funding may be readily available, but Elie Bleier be put into making this process more than capital will be required for Bleier expects to complete his M.A. in human-centric. Human-centricity is success. Partnering with healthcare Philosophy from Tel Aviv University and holds a B.A. in Economics-Philosophy of extreme value within healthcare organizations, and building on our from Columbia University. settings, which are typically centered model for change as outlined above, can around outcomes (i.e., curing specific help instantiate innovation, benefitting Seed startups need to move quickly illnesses). In complement, digital startups, their partner organizations, health startups should aim towards and even populations at large. to develop a rough idea of how to improving the patient and provider solve unmet needs. But too often experience, keeping them front-and- The initial innovative spark can come center during solution building. from organizations external to the they spend months working in isolation startup or healthcare organization. To optimize this human-centric And while partnerships are not quick on a prototype which doesn’t meet process, a wide range of professions and easy nor a guarantee of success, real world needs can be invoked. While this they tend to challenge the rigidity of should include strategic design the silos which separate technologists professionals able to facilitate human- from their eventual contexts, thus centric methodologies, it can also improving outcome measured by include various other professions, the impact and scalability of any ranging from UX designers to given innovation. As we build the sociologists and beyond, to examine next generation of healthcare, we interactions between patients, hope that startups and healthcare families, and medical staff. organizations alike will recognize the mutual benefit in one another, and partner sooner rather than later.

76 COLLER VENTURE REVIEW 77 Coller Entrepreneurial Team Formation Venture Digest Age and High-Growth This paper studies startups systematically in the U.S. and finds that successful Entrepreneurship entrepreneurs are middle-aged, not young. In the U.S., the mean age at founding for the 1-in-1,000 fastest growing new ventures is 45. These findings are consistent with theories in which key entrepreneurial resources (such as human Azoulay, Pierre; Benjamin F. Jones; J. Daniel Kim; Javier Miranda capital, financial capital, and social capital) accumulate with age. Mechanisms American Economic Review: Insights, by which young people are proposed to have advantages (such as energy or April 2019 originality) may still be operating, but if so they appear to be overwhelmed

https://www.kellogg.northwestern.edu/ by other forces. The findings are similar when considering high-technology faculty/jones-ben/htm/Age%20and%20 sectors, entrepreneurial hubs, and successful firm exits. Prior experience in the High%20Growth%20Entrepreneurship.pdf specific industry predicts much greater rates of entrepreneurial success. This is a strong encouragement that it is never too late to become an entrepreneur!

Recommended by Prof. Francesca Cornelli, Kellogg School of Management or the first time, we separately Immigration and This is an important paper in the present environment: it uses U.S. introduce our Venture Digest, Entrepreneurship administrative data, a representative sample, and Fortune 500 data to study referring our readers to some in the United States the role of immigrants in entrepreneurship. Across all three data sets, the F authors find that immigrants present a “right shift” in new venture formation, where immigrants tend to start more firms of each size per member of their of the year’s best reads in venture, Azoulay, Pierre; Benjamin F. Jones; population. It looks at how often immigrants start companies, how many jobs as suggested by our Advisory Board. J. Daniel Kim; Javier Miranda Working Paper 27778 these firms create, and how firms founded by native-born individuals compare. National Bureau of Economic Research, The findings suggest that immigrants appear to play a relatively strong role These articles cross the gamut from September 2020 in expanding labor demand relative to labor supply, compared to the native-

Entrepreneurial Team Formation to https://www.kellogg.northwestern.edu/ born population, that immigrants act more as “job creators” than “job takers,” faculty/jones-ben/htm/Immigration_and_ and that they play outsized roles in U.S. high-growth entrepreneurship. Funding New Ventures, Leadership Entrepreneurship_in_the_United_States.pdf Recommended by Prof. Francesca Cornelli, in Venture, Social Entrepreneurship, Kellogg School of Management Success in Venture Creation, and Entrepreneurial Entrepreneurial team formation—the process through which founders Change in Private Equity. Team Formation establish a team to start a new venture—has important implications for team performance and entrepreneurial success. This article reviews entrepreneurial teams and how they play a key role in investment decisions Our digest will continue to be Lazar, Moran; Ella Miron-Spektor; updated, and we are pleased to Rajshree Agarwal; Miriam Erez; and overall venture success. Investors often bet on the “jockey” (i.e., the Brent Goldfarb; Gilad Chen team) rather than on the “horse” (i.e., the idea). But how do these teams come provide hard copies upon request. Academy of Management Annals, about in the first place? This article reviews recent work on entrepreneurial January 2020 team formation and offers a useful framework for entrepreneurs searching https://www.researchgate.net/ for co-founders based on an integrative model that considers the evolving publication/334119413_Entrepreneurial_ entrepreneurial context and the opportunity for teams to integrate research Team_Formation from different disciplines (e.g., economics, psychology, and sociology).

Recommended by Prof. Ella Miron-Spektor, INSEAD

78 COLLER VENTURE REVIEW 79 Funding New Leadership Ventures in Venture

Sizing Up Entrepreneurial In this forthcoming Academy of Management Journal article, Professor Both/And Leadership In this article, Smith, Lewis, and Tushman describe contemporary paradoxes Potential: Gender Differences Laura Huang (HBS) and colleagues uncover the subtle mechanism leading of leadership and effective ways to cope with them. They propose a new model— to gender inequality in startup funding. Specifically, they take a deep dive one in which the goal of leadership is to maintain a dynamic equilibrium in the in Communication and Smith, Wendy K.; Marianne W. Lewis; Investor Perceptions of Long- into founder-investors interactions and find that male founders use more Michael L. Tushman organization. This is at the center of paradoxical leadership. By contrast, trying abstract communication which is perceived as growth-oriented and thus HBR Magazine, May 2016 to shift the hearts and minds of senior team members is challenging and time- Term Growth and Scalability more likely to secure funding. Women communicate more concretely than consuming. Moreover, their roles and responsibilities often lead senior people https://hbr.org/2016/05/both-and-leadership men which the audience described as psychologically close. Differences in to deeply identify with one goal or another, fostering conflict. To unleash the Huang, Laura; Priyanka Joshi; linguistic abstraction are the result of several interrelated processes—including power of paradox, therefore, leaders must build supporting organizational Cheryl Wakslak; Andy Wu but not limited to social network size and homogeneity, communication competencies into their senior team. Executives with this goal do not focus Academy of Management Journal, April 2020 motives involving seeking proximity or distance, perceptions of audience on being consistent; instead they purposefully and confidently embrace the homogeneity and distance, as well as experience of power. Using a multi- paradoxes they confront. By adopting this paradoxical mindset, leaders can https://journals.aom.org/doi/10.5465/ method study of thousands of pitch decks and a clever experimental balance competing demands and promote innovation and sustainability. amj.2018.1417 design, the authors show this is a widespread phenomenon and further This requires managers to both separate and connect opposing forces. document the negative impact on funding among female-led startups. Recommended by Prof. Ella Miron-Spektor, Recommended by Prof. Gary Dushnitsky, INSEAD London Business School

We Ask Men to Win and This article reviews why male entrepreneurs tend to raise higher levels of Social Women Not to Lose: Closing funding relative to their female counterparts. This research suggests that the Gender Gap in Startup the gap originates with a gender bias in the questions that investors pose to Entrepreneurship Funding entrepreneurs. Investors tend to ask male entrepreneurs promotion-focused questions and female entrepreneurs prevention-focused questions. The paper offers evidence regarding tactics that can be employed to diminish Kanze, Dana; Laura Huang; the gender disadvantage in funding outcomes. Specifically, the authors find Mark A. Conley; E. Tory Higgins Academy of Management Journal Vol. 61, that entrepreneurs can significantly increase funding for their startups when A Calculation of the Social The authors in this paper estimate the social returns to investments in responding to prevention-focused questions with promotion-focused answers, No. 2, April 2018 Returns to Innovation innovation, and suggest that the social returns are very large. Even under very which has theoretical as well as practical implications for entrepreneurship. https://journals.aom.org/doi/10.5465/ conservative assumptions, it is difficult to find an average return below $4 per $1 spent. Accounting for health benefits, inflation bias, or international spillovers amj.2016.1215 Recommended by Prof. Ella Miron-Spektor, Jones, Benjamin F; INSEAD Lawrence H. Summers can bring the social returns to over $20 per $1 spent, with internal rates of Working Paper 27863 return approaching 100%. The implication is that innovation investments can National Bureau of Economic Research, credibly raise economic growth rates and extend lives, paying for their costs September 2020 many times over. And because the social returns exceed the private returns, https://www.kellogg.northwestern. public policy has a central role, and opportunity, in unleashing these gains. edu/faculty/jones-ben/htm/Social%20 Recommended by Prof. Francesca Cornelli, Returns%20_%20Jones%20Summers%20_%20 Kellogg School of Management NBER.pdf

80 COLLER VENTURE REVIEW 81 Success in Systematic Change Venture Creation in Private Equity

Predicting Success in the In this article, Dr. Moreno Bonaventura and colleagues use big data methodology Mutual Funds as Venture In this paper, the authors present evidence to suggest that open-end mutual Worldwide Startup Network to predict startup success. They use a highly visible signal—the professionals Capitalists? Evidence funds investing in private companies are increasingly more likely to invest in (i.e., founders, employees, advisors, investors, and board members) that from Unicorns unicorns. For example, the aggregate holdings of open-end mutual funds in startups have in common—to calculate centrality measures for more than unicorns has increased from $1 billion to more than $8 billion from 2010-2016. Bonaventura, Moreno, 40,000 startups across 117 countries. Sociological and economic research has In addition, rounds with mutual fund participation have stronger redemption Ciotti, V., Panzarasa, P. et al Chernenko, Sergey; Joshua Lerner; Sci Rep 10, 345, January 2020 previously investigated the impact of knowledge spillovers, involvement in and IPO-related rights and less board representation, all of which improve Yao Zeng inter-firm alliances, and network position on firms’ performance, innovation the liquidity of the underlying securities. This affects the governance of https://doi.org/10.1038/s41598-019-57209-w The Review of Financial Studies, hhaa100, capacity, propensity to collaborate, and growth rates. Yet, whether the centrality September 2020 entrepreneurial firms and the evolving dynamics of investment in private firms.

in the professional network of newly established knowledge-intensive firms https://doi.org/10.1093/rfs/hhaa100 Recommended by Prof. Joshua Lerner, can help predict their long-term economic success has largely remained a Harvard Business School moot question. This paper is among the first attempts to pave the way in this direction, and represents a contribution, from a different angle. This work is intended to elucidate the role that network mechanisms might play in sustaining success, with the authors finding that a startup’s centrality score is a significant predictor of the likelihood of a successful liquidity event.

Recommended by Prof. Gary Dushnitsky, London Business School

Private Portfolio In this paper the authors use the comprehensive Burgiss data to approximate Value Creation in In this paper the authors use a proprietary EBRD database to measure value Attribution Analysis fund performance attributions at the portfolio level, and propose a method Private Equity creation at the level of the individual portfolio company (i.e., deal level). for providing attribution analysis to a portfolio of closed-end drawn- This is significant because while value creation and performance measurement down funds such as private equity buyout funds and venture capital funds. of private assets has always been a central yet contentious topic, it is not Brown, Gregory W.; Frank Ethridge; Biesinger, Markus; Cagatay Bircan; Tyler Johnson; Thomas Keck The method isolates performance attributes related to fund strategy, fund Alexander Ljungqvist straightforward to define the right benchmark within the broader asset class. June 2020 geography, commitment timing, and commitment sizing. A residual EBRD Working Paper No. 242 Statistical methodologies such as those discussed here which decompose

unexplained component can be viewed as a fund selection attribute. While Swedish House of Finance Research Paper fund performance to its components (attribution analysis) go deep splitting the https://papers.ssrn.com/sol3/papers. No. 20-17, September 2020 cfm?abstract_id=3624399 these confidence intervals are specific to the time periods used in the analysis, performance to many components based on the P&L, cash flows and market they provide a reasonable gauge for understanding what thresholds could be https://papers.ssrn.com/sol3/papers. multiple firms. The authors conclude that successful execution appears to be considered a significant skill for each attribute over a fairly long history. cfm?abstract_id=3607996 a key driver of investor returns, especially in growth, buyout, and secondary deals. Much like combination therapies in the treatment of certain diseases, Recommended by Prof. Eli Talmor, London Business School investor returns depend on how strategies are combined. Company operations and profitability improve in ways consistent with the successful implementation of value creation plans, and these improvements persist beyond PE funds’ exit.

Recommended by Prof. Eli Talmor, The Deregulation of the This article suggests that the notion of going public as a major step toward London Business School Private Equity Markets expanding and hiring may not be as big a deal as anticipated. The deregulation and the Decline in IPOs of the private equity markets—and in particular the National Securities Markets Improvement Act (NSMIA) of 1996—has made it possible for both VC-backed startups and the funds investing in them to raise large sums of private capital. Ewens, Michael; Joan Farre-Mensa The Review of Financial Studies, Together with the growing role of non-traditional investors such as PE or mutual hhaa053, May 2020 funds in the entrepreneurial finance market, this has helped bring about a new equilibrium where fewer startups go public, and those that go public are https://doi.org/10.1093/rfs/hhaa053 older. The authors show that the bargaining power of startup founders vis-`a- https://papers.ssrn.com/sol3/papers. cfm?abstract_id=3624399 vis investors has increased and that founders are using their increased control over exit decisions to stay private longer. How this new equilibrium is affecting the incentives and returns of startup investors remains an open question.

Recommended by Prof. Joshua Lerner, Harvard Business School

82 COLLER VENTURE REVIEW 83 Advisory Board

Prof. Gad Allon Professor Gad Allon is the Jeffrey Early-Career Research Award” of Prof. Francesca Professor Francesca Cornelli is the the board of editors of the Review of A. Keswin Professor and Professor the Production and Operations Dean of Northwestern University’s Economic Studies and as an associate of Operations, Information and Management Society. He is the Cornelli Kellogg School of Management. She editor at the Journal of Finance. She Decisions, and the Director of the Operations Management Department is also a Professor of Finance and holds is a research fellow at the Center for Management and Technology Program Editor of Management Science and the Donald P. Jacobs Chair of Finance. Economic and Policy Research, and at the University of Pennsylvania. serves on the editorial board previously served as a director of the Previously, she was Professor of of several journals. American Finance Association. Professor Allon’s research interests finance and Deputy Dean at the include operations management in Professor Allon is the Co-founder London Business School. She directed In January 2016 Professor Cornelli general, and service operations and of ForClass, a platform that enables and advanced the Private Equity helped create and became a board operations strategy in particular. professors to drive higher student Institute of London Business School, member of AFFECT, a committee He has been studying models of engagement and accountability in building a bridge between academia of the American Finance Association information sharing among firms their classrooms. He regularly consults and practice by partnering with designed to promote the advancement and customers both in service and firms both on service strategy and private equity leaders in London, of women academics in the field retail settings, as well as competition operations strategy. alumni and top academics in the field. of finance. Jeffrey A. Keswin Professor models in the service industry. His Dean, Kellogg School Professor Allon holds a Ph.D. in Professor Cornelli’s research interests Professor Cornelli holds an M.A. and Professor of Operations, articles have appeared in Management of Management, Management Science from Columbia include corporate governance, private and Ph.D. in Economics from Harvard Information and Decisions, Science, Manufacturing and Service Northwestern University Business School in New York and holds equity, privatization, bankruptcy, IPOs University and a B.A in Economics, The University of Pennsylvania Operations Management and a B.A. and M.A. from the Technion— and innovation policy. She has been summa cum laude, from Università Operations Research. Professor Allon Israel Institute of Technology. an editor of the Review of Financial Commerciale Bocconi. won the 2011 “Wickham Skinner Studies, and previously served on

Prof. Shai Professor Shai Bernstein is an Some of his latest research includes: Prof. Gary Professor Gary Dushnitsky is an Entrepreneurship Journal, Associate Professor in Entrepreneurial Does Venture Attract Human Capital Associate Professor of Strategy & and Nature Biotechnology. Bernstein Management at Harvard Business and The Creation of Evolution of Dushnitsky Entrepreneurship at the London Business Professor Dushnitsky serves as School and a Faculty Research Fellow Entrepreneurial Public Markets School. He serves as a Senior Fellow the Co-Editor of the Strategic at the National Bureau of Economic at The Mack Institute for Innovation Professor Bernstein holds a Ph.D. Entrepreneurship Journal. He received Research (NBER). His research Management at the Wharton School, in Business Economics from academic distinctions including the focuses on financial issues related University of Pennsylvania. Harvard University. 2013 SMS Emerging Scholar Award to startups and high growth firms, Professor Dushnitsky’s work focuses and the 2009 Kauffmann Junior and their interaction with innovation on the economics of entrepreneurship Faculty Fellowship, has been featured and entrepreneurial activity. Prior and innovation, and he advises in Business Week, CNBC, Dow Jones to joining Harvard Business School, corporations in the Financial Industry, News, Entrepreneur Magazine, he was a faculty member at Stanford FMCG, Clean Tech, and Pharma Financial Times, and has participated Graduate School of Business. sectors. He explores the shifting at the YPO, World Economic Forum, landscape of entrepreneurial finance, OECD, EVCA, and BVCA. Associate Professor in Associate Professor of Strategy exploring such topics as corporate Professor Dushnitsky holds a Entrepreneurial Management, & Entrepreneurship, venture capital, crowdfunding, and Ph.D. in Strategy from New York Harvard Business School London Business School accelerators. His research appeared in University and a B.A. and M.Sc. leading academic journals, including from Tel Aviv University. Organization Science, Strategic Management Journal, Strategic

84 COLLER VENTURE REVIEW 85 Prof. Joshua Professor Joshua Lerner is the Jacob Innovation Policy and the Economy. Prof. Scott Stern Professor Scott Stern is the David In 2005 he was awarded the Kauffman H. Schiff Professor in Entrepreneurial He founded and runs the Private Sarnoff Professor of Management at Prize Medal for Distinguished Lerner Management at Harvard Business Capital Research Institute, a nonprofit the MIT Sloan School of Management Research in Entrepreneurship. School. His research focuses on devoted to encouraging access to data and a Professor of Technological Professor Stern works with practitioners venture capital and private equity and research, and has been a frequent Innovation, Entrepreneurship, in bridging the gap between academic organizations, particularly policies leader of and participant in the World and Strategic Management. He research and the practice of innovation on innovation and how they impact Economic Forum projects and events. was previously a Professor at the and entrepreneurship through advising firm strategies. He has authored Kellogg School of Management and He is the winner of Sweden’s Global startups and other growth firms in several books and publications Non-Resident Senior Fellow at the Entrepreneurship Research Award and the area of entrepreneurial strategy, including The Architecture of Brookings Institution. the Cheng Siwei Award for Venture as well as working with governments Innovation, The Comingled Code, Capital Research. Professor Stern’s research explores how and other stakeholders on policy issues Innovation and Its Discontents, innovation and entrepreneurship differ related to competitiveness and regional Boulevard of Broken Dreams, The Professor Lerner holds a Ph.D. in from traditional economic activities, performance. He is the director and Money of Invention, Patent Capital, Economics from Harvard University and the consequences of these co-founder of the Innovation Policy and The Venture Capital Cycle. and graduated from Yale College with Jacob H. Schiff Professor., David Sarnoff Professor differences for strategy and policy. His Working Group at the National Bureau a special divisional major. Entrepreneurial Management, Professor Lerner co-directs the of Management, MIT Sloan research in the economics of innovation of Economic Research. Harvard Business School National Bureau of Economic School of Management and entrepreneurship focuses on Professor Stern holds a Ph.D. in Research’s Productivity, Innovation, entrepreneurial strategy, innovation- Economics from Stanford University and Entrepreneurship Program and driven entrepreneurial ecosystems, and and a B.A. in Economics from New serves as co-editor of their publication, innovation policy and management. York University.

Prof. Ella Professor Ella Miron-Spektor is an Behavior and Human Decision Processes, Prof. Eli Talmor Professor Eli Talmor is an Emeritus Professor Talmor practices venture Associate Professor of Organizational and Journal of Applied Psychology. Professor of Accounting at the London capital and is a serial cornerstone Miron-Spektor Behavior at INSEAD. Her research Business School and Founder of its investor and the co-founder of Her work has been profiled in media focuses on personal and organizational Institute of Private Equity. He has SunRay Renewable Energy which was outlets such as Harvard Business factors that promote creativity, served on the board of Governors acquired in Europe’s largest venture Review, Forbes, CBS, and NBS news. learning, and entrepreneurial success. of London Business School, Tel exits of 2010. She co-organized several international She studies team characteristics that Aviv University and the advisory conferences, including the Academy He is the co-author of Private Capital, contribute to innovation and learning, board of the African Venture Capital of Management Specialized considered the reference book on the the formation of entrepreneurial Association. He was previously on the Conference: From-Start-up to Scale- private equity industry. teams, strategies that enable leaders finance faculty at the University of Up in 2018. She serves on the Editorial to cope with competing demands at California (UCLA and Irvine), Tel Aviv Professor Talmor holds a Ph.D. Review Board of Organization work, and the influence of cultural University and the Wharton School in Economics & Management from Science and as Guest Editor for diversity on creativity. of the University of Pennsylvania. the University of North Carolina at Organizational Behavior and Human Chapel Hill and a B.Sc., cum laude, Professor Miron-Spektor’s award- Decision Processes and Academy of from the Technion—Israel Institute Associate Professor of winning research studying factors Management Discoveries. Emeritus Professor of of Technology. Organizational Behavior, that contribute to team innovation Accounting and Founder Professor Miron-Spektor holds a Ph.D. INSEAD and learning has been published in of its Private Equity Institute, in Organizational Psychology from the top management journals, including London Business School Technion—Israel Institute of Technology the Academy of Management Journal, and completed a Postdoctoral Fellowship Organization Science, Organizational at Carnegie Mellon University.

86 COLLER VENTURE REVIEW 87 Editors

Dr. Leslie Dr. Leslie E. Broudo leads the Dr. Broudo holds an MBA Coller Institute of Venture at from the Wharton School of the E. Broudo Tel Aviv University. University of Pennsylvania in Management and Operations, She is a business professional and a Ph.D. in Anthropology from the anthropologist recognized for University of Pennsylvania, and a B.A. leading high-impact change at the in Political Science and Economics intersection of theory and practice. from Bryn Mawr College. Her public and private sector roles have spanned new and established technology ventures, private equity, and university entrepreneurship initiatives.

Head of the Coller Institute of Venture and Head of Sofaer Global MBA at the Coller School of Management, Tel Aviv University

Prof. Moshe Professor Moshe Zviran is Dean of Professor Zviran held academic positions the Coller School of Management at at the Naval Postgraduate School, Zviran Tel Aviv University, and a Professor of The Claremont Graduate University, Information Systems. He is the Isaac and Ben-Gurion University. Gilinsky Chair of Entrepreneurship, Professor Zviran holds an M.Sc. Technology, Innovation and and Ph.D. in Information Systems Management, and serves as the as well as a B.Sc. in Mathematics Academic Director of the Eli Hurvitz and Computer Science from Tel Institute for Strategic Management Aviv University. and the Coller Institute of Venture. Professor Zviran’s research interests include entrepreneurship and innovation, information and cyber security, and information systems Dean of the Coller planning and policy. He has published School of Management, numerous articles and authored two Tel Aviv University books on Information Systems. He is a consultant for leading organizations in Israel and serves as a board member in several companies and organizations.

88 COLLER VENTURE REVIEW In theory, theory and practice are the same. In practice, they are not.

Albert Einstein

The Editorial Board

Editor-in-Chief Prof. Moshe Zviran Dean, Coller School of Management Tel Aviv University

Managing Editor Dr. Leslie Broudo Head, Coller Institute of Venture Coller School of Management Tel Aviv University

Editorial Advisor Prof. Joshua Lerner Jacob H. Schiff Professor of Investment Banking Harvard Business School Harvard University

Editorial Offices Coller Institute of Venture Coller School of Management Tel Aviv University Ramat Aviv 6997801, Israel

Email: [email protected]

ISSN 2709-0582