2020 ■ VOLUME 68, No 1
CANADIAN TAX JOURNAL REVUE FISCALE CANADIENNE
PEER-REVIEWED ARTICLES The Evaluation of Job Tax Incentives: An Analysis of a Regional Tax Alessandro Zeli
POLICY FORUM Editor’s Introduction—Taxes and Spending in Canada’s 43rd Parliament Kevin Milligan Tax Reform in Canada’s 43rd Parliament—Politics, Policy, and Second-Best Choices Sean Speer Expenditures, Efficiency, and Distribution—Advice for Canada’s 43rd Parliament Rob Gillezeau and Trevor Tombe
SYMPOSIUM The Future of Work: The Gig Economy and Pressures on the Tax System Celeste M. Black Automation and Workers: Re-Imagining the Income Tax for the Digital Age Jinyan Li, Arjin Choi, and Cameron Smith Tim Edgar: The Accidental Comparatist Kim Brooks Moving to a More “Certain” Test for Tax Residence in Australia: Lessons for Canada? Michael Dirkis
AWARDS Douglas J. Sherbaniuk Distinguished Writing Award / Prix d’excellence en rédaction Douglas J. Sherbaniuk Canadian Tax Foundation Regional Student-Paper Awards / Prix régionaux du meilleur article par un étudiant de la Fondation canadienne de fiscalité Best Newsletter Article by a Young Practitioner Award / Prix pour le meilleur article de bulletin par un jeune fiscaliste Canadian Tax Foundation Lifetime Contribution Award / Prix de la Fondation canadienne de fiscalité pour une contribution exceptionnelle
(Continued on inside front cover) (Continued from outside front cover)
FEATURES Finances of the Nation: Survey of Provincial and Territorial Budgets, 2019-20 David Lin Current Cases: (FCA) Tedesco v. Canada; (TCC) Eyeball Networks Inc. v. The Queen; (UKUT) (TCC) Irish Bank Resolution Corporation Ltd (in Special Liquidation) and Irish Nationwide Building Society v. Revenue and Customs Brian Studniberg, Maressa Singh, Michael Templeton, and Joel Nitikman Personal Tax Planning / Planification fiscale personnelle : Due Diligence Defence to Liability for Unpaid Statutory Remittances / Défense de diligence raisonnable relativement à la responsabilité des versements obligatoires impayés Wayne D. Gray Corporate Tax Planning: GAAR: An Economic Test?— The Courts Divide Brian R. Carr, Brittany Finn, and Ryan Wolfe Current Tax Reading Alan Macnaughton and Jinyan Li ■ CANADIAN TAX JOURNAL EDITORIAL BOARD/ COMITÉ DE RÉDACTION DE LA REVUE FISCALE CANADIENNE
■ Editors/Rédacteurs en chef Alan Macnaughton Daniel Sandler University of Waterloo EY Law llp Kevin Milligan University of British Columbia
■ Practitioners/Fiscalistes ■ University Faculty/Universitaires Brian J. Arnold Reuven Avi-Yonah Tax Consultant, Toronto University of Michigan Thomas A. Bauer Richard M. Bird Bennett Jones llp, Toronto University of Toronto Stephen W. Bowman Robin W. Boadway Bennett Jones llp, Toronto Queen’s University C. Anne Calverley Neil Brooks Dentons Canada llp, Calgary York University R. Ian Crosbie Arthur Cockfield Davies Ward Phillips & Vineberg llp, Toronto Queen’s University Cy M. Fien Graeme Cooper Fillmore Riley llp, Winnipeg University of Sydney James P. Fuller Bev G. Dahlby Fenwick & West llp, Mountain View, ca University of Calgary Edwin C. Harris James B. Davies McInnes Cooper, Halifax University of Western Ontario William I. Innes David G. Duff Rueter Scargall Bennett llp , Toronto University of British Columbia Brent Perry Judith Freedman Felesky Flynn llp, Calgary Oxford University Scott Jeffery Vijay Jog KPMG llp, Vancouver Carleton University Howard J. Kellough Jonathan R. Kesselman Davis llp, Vancouver Simon Fraser University Heather Kerr Ernst & Young llp/Couzin Taylor llp, Toronto Kenneth J. Klassen University of Waterloo Edwin G. Kroft Bennett Jones llp, Vancouver Gilles N. Larin Elaine Marchand Université de Sherbrooke Banque Nationale du Canada, Montréal Amin Mawani Janice McCart York University Blake Cassels & Graydon llp, Toronto Jack Mintz Thomas E. McDonnell University of Calgary Toronto Martha O’Brien Matias Milet University of Victoria Osler Hoskin & Harcourt llp, Toronto Suzanne Paquette W. Jack Millar Université Laval Millar Kreklewetz llp, Toronto Abigail Payne Michael J. O’Connor McMaster University Sunlife Financial Inc., Toronto Michael R. Veall François Vincent McMaster University KPMG Law llp, Chicago www.ctf.ca/www.fcf-ctf.ca Call for Book Proposals
The Canadian Tax Foundation, an independent, not-for-profit research and educational organization, is seeking proposals for books in the areas of taxation and public finance. Since its inception in 1945, the Foundation has published many books and articles on a wide range of subjects within its areas of interest. The Foundation seeks proposals for research projects that will
n result in a book on a single topic of interest in the area of taxation or public finance; n be undertaken by an experienced researcher who has expertise in an area of taxa - tion or public finance; and n be carried out within a time frame that is reasonable, given the nature of the project.
Projects selected by the Foundation may qualify for its full or partial financial support of the research and for its underwriting of the publication costs. The Foundation retains the absolute right at its sole discretion to choose whether to support a given proposal or to publish a project. Interested parties should send a brief written outline of a proposal, for initial consider - ation by the Foundation, to:
Heather Evans Executive Director and Chief Executive Officer Canadian Tax Foundation/Fondation canadienne de fiscalité 145 Wellington Street West, Suite 1400 Toronto, Ontario M5J 1H8 [email protected]
For further information, please contact the director, as indicated above, or the co-chairs of the Canadian Tax Foundation Research Committee:
Hugh Woolley c/o Canadian Tax Foundation/Fondation canadienne de fiscalité
Kim Brooks c/o Canadian Tax Foundation/Fondation canadienne de fiscalité
■ i ■ Appel de propositions de livres
La Fondation canadienne de fiscalité (FCF ) / Canadian Tax Foundation, un organisme sans but lucratif indépendant de recherche et à caractère éducatif, souhaite recevoir des propo - sitions de livres dans les domaines de la fiscalité et des finances publiques. Depuis sa fondation en 1945, la FCF a publié de nombreux livres et articles sur divers sujets dans ses champs d’intérêt. La FCF souhaite obtenir des propositions de projets de recherche qui :
n mèneront à la rédaction d’un livresur un sujet unique d’intérêt en fiscalité ou en finances publiques; n seront dirigés par un chercheur chevronné ayant une expertise dans un domaine de la fiscalité ou des finances publiques; n seront effectués dans un délai raisonnable, compte tenu de la nature du projet.
Les projets qui seront sélectionnés par la FCF pourront être partiellement ou totalement admissibles à une aide financière pour la recherche et les frais de publication. La FCF se réserve le droit absolu, et à sa seule discrétion, d’appuyer une proposition particulière ou de publier un projet. Toute personne intéressée doit faire parvenir un bref sommaire de la proposition pour examen initial par la FCF à :
Heather Evans Directrice exécutive et chef de la direction Canadian Tax Foundation/Fondation canadienne de fiscalité 145 Wellington Street West, Suite 1400 Toronto, Ontario M5J 1H8 [email protected]
Pour plus d’information, veuillez communiquer avec le directeur, tel qu’il est mentionné plus haut, ou avec les co-présidentes du comité de recherche de la Fondation canadienne de fiscalité :
Hugh Woolley a/s Canadian Tax Foundation/Fondation canadienne de fiscalité
Kim Brooks a/s Canadian Tax Foundation/Fondation canadienne de fiscalité
■ ii ■ ■ Recent and Upcoming Events * ■ Activités récentes et à venir *
We remain optimistic that most of our core conferences will proceed, particularly those scheduled for the fall. However, given the protocols to which many of our volunteer program committee members are currently subject, and the uncertainty of travel over the coming weeks, we will be holding most of the planning meetings virtually. We will rely on our live webcasts as an option for delegates, and we’ll be prepared to postpone or modify our events as appropriate. Please check our events page to see which events have been can - celled or postponed.
Nous demeurons optimistes que la plupart de nos conférences principales procèderont tel que prévu, particulièrement celles prévues à l’automne. Nous tiendrons toutefois la plupart de nos réunions de planification de façon virtuelle et à distance et ce, par respect pour les protocoles auxquels nos membres de comités bénévoles de planification des conférences sont présentement soumis et quant à l’incertitude des déplacements au cours des prochaines semaines. À titre d’option pour nos participants, nous compterons sur nos webémission en direct et nous demeurons à l’écoute pour reporter ou modifier nos événe - ments, le cas échéant. Veuillez svp consulter notre page événements afin de connaitre quels événements ont été annulés ou reportés.
■ iii ■ © 2020, Canadian Tax Foundation/Fondation canadienne de fiscalité
Disclaimer. The material contained in this publication is not intended to be advice on any particular matter. No subscriber or other reader should act on the basis of any matter contained in this publication without considering appropriate professional advice. The publisher, and the authors and editors, expressly disclaim all and any liability to any person, whether a purchaser of this publication or not, in respect of anything and of the consequences of anything done or omitted to be done by any such person in reliance upon the contents of this publication. Opinions expressed by individual writers are not necessarily endorsed by the Canadian Tax Foundation and its members. Photocopying and reprinting. Permission to photocopy or reprint any part of this publication for distribu- tion must be applied for in writing; e-mail: [email protected]. Advertising. Inquiries relating to advertisements should be directed to Christine Escalante, e-mail: [email protected]. Exonération de responsabilité. Le contenu de cette publication ne doit être interprété d’aucune façon comme un avis ou une opinion. L’abonné ou le lecteur ne devrait pas fonder ses décisions sur le contenu de cette publication sans envisager une consultation professionnelle appropriée. L’éditeur et les auteurs réfutent toute responsabilité envers toute personne, qu’elle soit abonnée ou non, rela- tivement à toute conséquence résultant d’actes ou omissions faits en fonction du contenu de la présente publication. Les opinions exprimées par les auteurs particuliers ne sont pas nécessairement appuyées par la Fondation canadienne de fiscalité et ses membres. Photocopie et réimpression. L’autorisation de photocopier ou de réimprimer toute portion de cette publication à des fins de distribution devra être obtenue en adressant une demande écrite à permissions @ctf.ca. Annonces publicitaires. Toutes demandes concernant les annonces publicitaires devront être adres- sées à Christine Escalante, courriel : [email protected].
Canadian Tax Journal Revue fiscale canadienne Published four times per year Publiée quatre fois l’an Price: $75 per copy (plus applicable taxes) Prix : 75 $ l’exemplaire (taxes en sus) Subscription rate: $343.75 per year (plus Abonnement : 343,75 $ par an (taxes en sus) applicable taxes) (Numéro d’enregistrement de TVH : (HST registration no. R-106867260) R‑106867260)
Canadian Tax Foundation La Fondation canadienne de fiscalité 145 Wellington Street West 1250, boul. René-Lévesque ouest Suite 1400 Bureau 2935 Toronto, Canada M5J 1H8 Montréal (Québec) H3B 4W8 Telephone: 416-599-0283 Téléphone : 514-939-6323 Toll Free: 1-877-733-0283 Télécopieur : 514-939-7353 Facsimile: 416-599-9283 Internet : www.fcf-ctf.ca Internet: www.ctf.ca
2020, vol. 68, no. 1 2020, vol. 68, no 1 (Issued April 2020) (publication : avril 2020)
ISSN 0008-5111 ISSN 0008-5111
Printed in Canada Imprimée au Canada 6,200 04-20 6,200 04-20
■ iv ■ ■ 2020 VOLUME 68, No 1
Canadian Tax Journal Revue fiscale canadienne
PEER-REVIEWED ARTICLES 1 The Evaluation of Job Tax Incentives: An Analysis of a Regional Tax ALESSANDRO ZELI
POLICY FORUM 33 Editor’s Introduction—Taxes and Spending in Canada’s 43rd Parliament KEVIN MILLIGAN
35 Tax Reform in Canada’s 43rd Parliament—Politics, Policy, and Second-Best Choices SEAN SPEER
49 Expenditures, Efficiency, and Distribution— Advice for Canada’s 43rd Parliament ROB GILLEZEAU AND TREVOR TOMBE
SYMPOSIUM 69 The Future of Work: The Gig Economy and Pressures on the Tax System CELESTE M. BLACK
99 Automation and Workers: Re-Imagining the Income Tax for the Digital Age JINYAN LI, ARJIN CHOI, AND CAMERON SMITH
125 Tim Edgar: The Accidental Comparatist KIM BROOKS
143 Moving to a More “Certain” Test for Tax Residence in Australia: Lessons for Canada? MICHAEL DIRKIS
AWARDS 169 Douglas J. Sherbaniuk Distinguished Writing Award / Prix d’excellence en rédaction Douglas J. Sherbaniuk
171 Canadian Tax Foundation Regional Student-Paper Awards / Prix régionaux du meilleur article par un étudiant de la Fondation canadienne de fiscalité
179 Best Newsletter Article by a Young Practitioner Award / Prix pour le meilleur article de bulletin par un jeune fiscaliste
183 Canadian Tax Foundation Lifetime Contribution Award / Prix de la Fondation canadienne de fiscalité pour une contribution exceptionnelle
■ v ■ FEATURES 185 Finances of the Nation: Survey of Provincial and Territorial Budgets, 2019-20 DAVID LIN
251 Current Cases: (FCA) Tedesco v. Canada; (TCC) Eyeball Networks Inc. v. The Queen; (UKUT) (TCC) Irish Bank Resolution Corporation Ltd (in Special Liquidation) and Irish Nationwide Building Society v. Revenue and Customs BRIAN STUDNIBERG, MARESSA SINGH, MICHAEL TEMPLETON, AND JOEL NITIKMAN
281 Personal Tax Planning: Due Diligence Defence to Liability for Unpaid Statutory Remittances WAYNE D. GRAY
313 Planification fiscale personnelle : Défense de diligence raisonnable relativement à la responsabilité des versements obligatoires impayés WAYNE D. GRAY
351 Corporate Tax Planning: GAAR: An Economic Test?— The Courts Divide BRIAN R. CARR, BRITTANY FINN, AND RYAN WOLFE
391 Current Tax Reading ALAN MACNAUGHTON AND JINYAN LI
■ vi ■ ■ Canadian Tax Journal
The Canadian Tax Journal publishes research in, and informed comment on, taxation and public finance, with particular relevance to Canada. To this end, the journal invites interested parties to submit manuscripts for possible publication as peer-reviewed articles, and it especially welcomes work that contributes to the analysis, design, and implementation of tax policies. Articles may be written in English or French and should present an original analysis of the topic. Submitted work, or any substantial part or version thereof, must not have been previously published, either in print or online, and it must not be submitted or scheduled for publication elsewhere. The journal welcomes shorter submissions (from 4,000 to 8,000 words) focused on specific topics as well as longer submissions (to a maximum of 20,000 words) that analyze issues in depth. Submitted articles are subject to a double-blind peer review; authors’ identities are not known to reviewers, and reviewers’ identities are not known to authors. (Non-peer-reviewed contributions may appear elsewhere in the journal.) Final decisions on publication of articles are made by the editors, Alan Macnaughton, Daniel Sandler, and Kevin Milligan, on the advice of reviewers. Many reviewers are drawn from the editorial board (listed on the inside front cover of this journal), although ad hoc reviewers are also consulted. Submissions may be (1) accepted outright; (2) accepted if recommended revisions are made; (3) revised by the authors, as requested by the editors on the advice of reviewers, and resubmitted for further review; or (4) rejected with reasons. The time from submission to the first edi - torial decision is usually two months or less. Prospective contributors should submit a copy of the manuscript to the journal’s editorial department. The preferred method of submission is by e-mail with an attached Word docu- ment. E-mail inquiries are welcome: write to [email protected]. Contributors are responsible for providing complete and accurate citations to sources, a detailed abstract (200 to 400 words), and up to six keywords for indexing purposes. The full text of many articles that have appeared in the Canadian Tax Journal since 1991 can be found on the Canadian Tax Foundation’s website: www.ctf.ca. Additionally, the journal in its entirety appears in the Canadian Tax Foundation’s TaxFind, which is updated regularly. The Canadian Tax Journal is indexed in EconLit, ABI Inform, LegalTrac, Index to Canadian Legal Literature, CCH Canadian’s Canadian Income Tax Research Index, Carswell’s Income Tax References, Accounting and Law Index, Current Law Index, Canadian Index, Canadian Periodicals Index, Index to Canadian Legal Periodical Literature, Index to Legal Periodicals and Books, and PAIS International in Print.
■ vii ■ ■ Revue fiscale canadienne
La Revue fiscale canadienne publie des recherches et des commentaires éclairés sur la fiscalité et les finances publiques, particulièrement pertinents pour le Canada. À cette fin, la revue invite les personnes intéressées à soumettre des articles en vue d’une éventuelle publication en tant qu’articles revus par des pairs, et elle accueille tout particulièrement les travaux qui contribuent à l’analyse, à la conception et à la mise en oeuvre des politiques fiscales. Les articles peuvent être rédigés en anglais ou en français et doivent présenter une analyse originale du sujet. Les articles soumis, ou toute partie substantielle ou version des articles, ne doivent pas avoir été publiés antérieurement en format papier ou électronique, et ne doivent pas être soumis ou prévus pour publication ailleurs. Vous pouvez soumettre pour publication, dans la revue fiscale, des articles plus courts (4 000 à 8 000 mots) sur des sujets particuliers ainsi que des articles plus longs (maximum de 20 000 mots) analysant des sujets en profondeur. Les articles soumis sont sujets à une double revue à l’aveugle par des pairs; l’identité des auteurs n’est pas connue des réviseurs et celle des réviseurs n’est pas connue des auteurs (certains articles non soumis à cette révision par des pairs peuvent paraître ailleurs dans la revue.) La décision finale de publier ou non un article est celle des rédacteurs en chef Alan Macnaughton, Daniel Sandler et Kevin Milligan, à la recommandation des réviseurs. Bien que certains réviseurs ad hoc soient aussi consultés, la majorité des réviseurs sont choisis parmi les membres du Comité de rédaction (énumérés à l’endos de la page couver - ture de la revue). Les articles soumis peuvent être 1) acceptés d’emblée; 2) acceptés après modifications; 3) modifiés par les auteurs tel que demandé par les rédacteurs en chef sur l’avis des réviseurs, et resoumis à une nouvelle révision; ou 4) rejetés avec raisons. Le temps écoulé entre la soumission d’un article et la première décision éditoriale est habitu - ellement de deux mois ou moins. Les aspirants contributeurs doivent soumettre un exemplaire de l’article proposé au service éditorial. Il est préférable que la soumission se fasse par courriel, avec une pièce jointe en Word. Les demandes de renseignements par courriel sont les bienvenues. Elles doivent être adressées à [email protected]. Les contributeurs doivent soumettre l’ensemble de leurs sources, un précis détaillé de leurs articles (entre 200 et 400 mots), et jusqu’à six mots clés aux fins d’indexation. On peut trouver le texte intégral de nombreux articles publiés dans la Revue fiscale canadienne depuis 1991 sur le site Internet de le Fondation : www.fcf-ctf.ca. De plus, la revue dans son entier se trouve dans TaxFind, qui est mis à jour régulièrement. La Revue fiscale canadienne est indexée sous EconLit, ABI Inform, LegalTrac, Index to Canadian Legal Literature, Canadian Income Tax Research Index de CCH Canadian, Income Tax References de Carswell, Accounting and Law Index, Current Law Index, Canadian Index, Canadian Periodicals Index, Index to Canadian Legal Periodical Literature, Index to Legal Periodicals and Books, et PAIS International in Print.
■ viii ■ ■ Canadian Tax Foundation
The Canadian Tax Foundation is Canada’s leading source of insight on tax issues. The Foundation promotes understanding of the Canadian tax system through analysis, research, and debate, and provides perspective and impartial recommendations concerning its equity, efficiency, and application. The Canadian Tax Foundation is an independent tax research organization and a regis - tered charity with over 12,000 individual and corporate members in Canada and abroad. For more than 70 years, it has fostered a better understanding of the Canadian tax system and assisted in the development of that system through its research projects, conferences, publications, and representations to government. Members find the Foundation to be a valuable resource both for the scope and depth of the tax information it provides and for its services, which support their everyday work in the taxation field. Government policy makers and administrators have long respected the Foundation for its objectivity, its focus on current tax issues, its concern for improvement of the Canadian tax system, and its significant contribution to tax and fiscal policy.
MEMBERSHIP Membership in the Foundation is open to all who are interested in its work. Membership fees are $399.00 a year, except that special member rates apply as follows: (a) $199.00 for members of the accounting and legal professions in the first three years following date of qualification to practise; (b) $199.00 for persons on full-time teaching staff of colleges, universities, or other educational institutions; (c) $40.00 for students in full-time attend - ance at a recognized educational institution; and (d) $171.00 for persons who have reached the age of 65 and are no longer actively working in tax. Memberships are for a period of 12 months dating from the receipt of application with the appropriate payment. Applications for membership are available from the membership administrator for the Canadian Tax Foundation: facsimile: 416-599-9283; Internet: www.ctf.ca; e‑mail: [email protected].
■ ix ■ ■ Fondation canadienne de fiscalité
La Fondation canadienne de fiscalité est un organisme indépendant de recherche sur la fiscalité inscrit sous le régime des œuvres de charité. Elle compte environ 12 000 membres au Canada et à l’étranger. Depuis plus de 70 ans, la FCF favorise une meilleure compréhension du système fiscal canadien et aide au développement de ce système par le biais de ses projets de recherche, conférences, publications et représentations auprès des gouvernements. Les membres considèrent l’étendue et le détail de l’information offerte par la FCF comme une importante ressource. Ils apprécient également les autres services de la FCF qui facilitent leur travail quotidien dans le domaine de la fiscalité. Les décideurs et administrateurs gouvernementaux respectent depuis longtemps l’objectivité de la FCF, son attention aux questions fiscales de l’heure, sa préoccupation envers l’amélioration du système fiscal canadien et son importante contribution au développement des politiques fiscales.
ADHÉSION Toute personne intéressée aux travaux de la FCF peut en devenir membre. Les droits d’adhésion sont de 399,00 $ par année, à l’exception des tarifs spéciaux suivants : a) 199,00 $ pour les personnes faisant carrière en comptabilité ou en droit pendant les trois premières années suivant leur admission à la profession; b) 199,00 $ pour le personnel enseignant à temps plein dans un collège, une université ou une autre maison d’enseignement; c) 40,00 $ pour les étudiants fréquentant à temps plein une maison d’enseignement reconnue; et d) 171,00 $ pour les personnes qui ont 65 ans et plus et qui ne travaillent plus activement en fiscalité. La période d’adhésion est de 12 mois, à compter de la réception de la demande accompagnée du paiement approprié. Il est possible de se procurer les demandes d’adhésion auprès de l’administratice responsible de l’adhésion à la FCF : télécopieur : 514-939-7353; Internet : www.fcf-ctf.ca; courriel : [email protected].
■ x ■ ■ BOARD OF GOVERNORS/CONSEIL DES GOUVERNEURS Elected December 1, 2019/Élu le 1 er décembre 2019
Albert Anelli, QC1* Ted Gallivan, ON3 Elizabeth Murphy, ON3 Cheryl Bailey, ON1 Rachel Gervais, ON1* Anu Nijhawan, AB2 Jeffery Blucher, NS2 Siobhan Goguen, AB2 Heather O’Hagan, ON3 Eoin Brady, ON1 Kay Gray, BC1 John Oakey, NS1 Mark Brender, QC2* Ken Hauser, BC2 Mitchell Sherman, ON2 Alycia Calvert, ON1* Soraya Jamal, BC2 Michael Smith, AB1 Marlene Cepparo, ON1* Timothy Kirby, AB2 Martin Sorensen, ON2* Grace Chow, ON1 Dean Landry, ON1* John Tobin, ON2 Allison Christians, QC3 Rick McLean, ON1 Dave Walsh, ON1 Michael Coburn, BC2 Stefanie Morand, ON2 Hugh Woolley, BC1 Marie-Claire Dy, BC2 Michael Munoz, AB3 Barbara Worndl, ON2* Olivier Fournier, QC2
* Executive Committee of the Board of Governors Comité de direction du conseil des gouverneurs 1 Nominee of the Chartered Professional Accountants of Canada 2 Nominee of the Canadian Bar Association 3 Non-sponsor
■ OFFICERS/MEMBRES DE LA DIRECTION Chair/Présidente du conseil Alycia Calvert Vice-Chair and Chair of the Executive Barbara Worndl Committee/Vice-présidente du conseil et présidente du comité de direction Second Vice-Chair/Deuxième vice-présidente Marlene Cepparo Past Chair/Président sortant du conseil Mark Brender Executive Director and Chief Executive Heather Evans Officer/Directrice exécutive et chef de la direction Director, Membership Development and Wayne Adams Community Relations/Directeur, Développement du programme de l’adhésion et relations avec la communauté Regional Director, Quebec/ Lucie Bélanger Directrice régionale du Bureau du Québec Director of Finance and Treasurer/ Shelly Ali Directrice financière et trésorière
■ STAFF/PERSONNEL Events and Web Manager/Directrice Roda Ibrahim des événements et du site Web Librarian/Bibliothécaire Judy Singh Managing Editor/Directeur de Michael Gaughan la rédaction
■ xi ■ ■ Canadian Tax Foundation Publications
The Foundation’s publications comprise a range of forms and delivery formats. A number of the regularly issued publications are distributed without charge to Foundation members: the Canadian Tax Journal (4 issues), Canadian Tax Highlights (12 issues, delivered elec- tronically), Tax for the Owner-Manager (4 issues, delivered electronically), Canadian Tax Focus (4 issues, delivered electronically), and the annual conference report. Monographs and books may be purchased on the Foundation’s website at www.ctf.ca.
Canadian Tax Journal — issued quarterly to members via www.ctf.ca (Non-Members $75 per copy, $343.75 per year). Newsletters Perspectives on Tax Law & Policy — issued quarterly to members via www.ctf.ca. Tax for the Owner-Manager — issued quarterly to members via www.ctf.ca. Canadian Tax Focus — issued quarterly; available to members and non-members via www.ctf.ca. Canadian Tax Highlights — archives (2000-2019) available to members via www.ctf.ca. Conference Reports — Reports of the proceedings of annual tax conferences (Members $40; Non‑Members $95). Latest issue: 2018 (Members $40; Non-Members $350). — Tax Dispute Resolution, Compliance, and Administration in Canada: Proceedings of the June 2012 Conference (Members $30; Non-Members $195) — Collections of papers delivered at regional and special tax conferences (British Columbia, Prairie Provinces, Ontario, and Atlantic Provinces) are available in USB format (Members $445; Non-Members $495). Finances of the Nation — Review of expenditures and revenues and some budgets of the federal, provincial, and local governments of Canada. PDFs for the years 2002-2012 are available on the CTF website at no cost. In 2014, “Finances of the Nation” began to appear as a feature in issues of the Canadian Tax Journal. Monographs 2019. Funding the Canadian City, Enid Slack, Lisa Philipps, Lindsay M. Tedds, and Heather L. Evans, eds. ($40 each) 2018. Tax Treaties After the BEPS Project: A Tribute to Jacques Sasseville, Brian J. Arnold, ed. (Members $60; Non-Members $90) 2018. Reforming the Corporate Tax in a Changing World, School of Public Policy of the University of Calgary (Members $30; Non-Members $50) 2017. Income Tax at 100 Years: Essays and Reflections on the Income War Tax Act, Jinyan Li, J. Scott Wilkie, and Larry F. Chapman, eds. (Members $60; Non-Members $90) 2016. Reform of the Personal Income Tax in Canada, School of Public Policy of the University of Calgary (Members $35; Non-Members $50) 2016. Canadian Taxation of Trusts, Elie S. Roth, Tim Youdan, Chris Anderson, and Kim Brown (Members $150; Non-Members $200; Students $50) 2016. User Fees in Canada: A Municipal Design and Implementation Guide, Catherine Althaus and Lindsay M. Tedds ($40 each) 2015. Timing and Income Taxation, 2d edition, Brian J. Arnold, Colin Campbell, Michael Hiltz, Richard Marcovitz, Shawn D. Porter, and James R. Wilson (Members $25; Non-Members $125; Students $25) 2015. Effective Writing for Tax Professionals, Kate Hawkins and Thomas E. McDonnell, QC (Members $35; Non-Members $40) 2014. After Twenty Years: The Future of the Goods and Services Tax, School of Public Policy of the University of Calgary (Members $25; Non-Members $35) 2013. Essays on Tax Treaties: A Tribute to David A. Ward, Guglielmo Maisto, Angelo Nikolakakis, and John M. Ulmer, eds. ($100 each)
■ xii ■ 2012. Tax Policy in Canada, Heather Kerr, Ken McKenzie, and Jack Mintz, eds. (Members $75; Non-Members $100; Students $50) 2011. Canadian Tax Foundation Style Guide, 5th edition (Members $35; Non-Members $40) 2011. International Financial Reporting Standards: Their Adoption in Canada, Jason Doucet, Andrée Lavigne, Caroline Nadeau, Jocelyn Patenaude, and Dave Santerre (Members $30; Non‑Members $40) 2011. Tax Expenditures: State of the Art — Selected Proceedings of the Osgoode 2009 Conference, Lisa Philipps, Neil Brooks, and Jinyan Li, eds. (Members $45; Non-Members $55; Students $30) 2010. Taxation of Private Corporations and Their Shareholders, 4th edition (Members $75; Non-Members $100; Students $25) Tax Professional Series (Please specify title and author when ordering.) 2003. The Taxation of Business Profits Under Tax Treaties, Brian J. Arnold, Jacques Sasseville, and Eric M. Zolt, eds. (softcover edition, $75) 2003. International Taxation in the Age of Electronic Commerce: A Comparative Study, Jinyan Li. Co-published with International Fiscal Association (Canadian Branch) (Members $95; Non-Members $145; Students $45) 1999. Countering Tax Treaty Abuses: A Canadian Perspective on an International Issue, Nathalie Goyette ($75 each) Canadian Tax Paper Series (Please specify publication number when ordering.) No. 112: 2009. Effective Responses to Aggressive Tax Planning: What Canada Can Learn from Other Jurisdictions, Gilles N. Larin and Robert Duong, with a contribution from Marie Jacques No. 111: 2009. Reforming Canada’s International Tax System: Toward Coherence and Simplicity, Brian J. Arnold (Members $100; Non-Members $125) No. 110: 2006. Financing Education and Training in Canada, 2d edition, Douglas Auld and Harry Kitchen No. 109: 2004. The Canadian Federal-Provincial Equalization Regime: An Assessment, Alex S. MacNevin No. 108: 2004. Venture Capital and Tax Incentives: A Comparative Study of Canada and the United States, Daniel Sandler No. 107: 2002. Municipal Revenue and Expenditure Issues in Canada, Harry M. Kitchen (Members $20; Non‑Members $40) No. 106: 2002. Taxes and the Canadian Underground Economy, David E.A. Giles and Lindsay M. Tedds No. 105: 2000. The Income Tax Treatment of Financial Instruments: Theory and Practice, Tim Edgar No. 104: 1999. Rationality in Public Policy: Retrospect and Prospect, A Tribute to Douglas G. Hartle, Richard M. Bird, Michael J. Trebilcock, and Thomas A. Wilson, eds. No. 103: 1999. Canadian Tax Policy, 3d edition, Robin W. Boadway and Harry M. Kitchen No. 102: 1997. Financing the Canadian Federation, 1867 to 1995: Setting the Stage, David B. Perry No. 101: 1997. General Payroll Taxes: Economics, Politics, and Design, Jonathan R. Kesselman No. 100: 1995. Growth of Government Spending in Alberta, Paul Boothe No. 99: 1995. Financing Education and Training in Canada, Harry Kitchen and Douglas Auld Special Studies in Taxation and Public Finance (Please specify publication number when ordering.) No. 2: 2000. Gambling and Governments in Canada, 1969-1998: How Much? Who Plays? What Payoff? François Vaillancourt and Alexandre Roy No. 1: 1998. Federal-Provincial Tax Sharing and Centralized Tax Collection in Canada, Ernest H. Smith
■ xiii ■ ■ Les publications de la Fondation canadienne de fiscalité
Les publications de la Fondation existent sous différentes formes et elles sont disponibles de diverses façons. Certaines de ces publications régulières sont distribuées gratuitement aux membres de la Fondation : la Revue fiscale canadienne (4 numéros), Faits saillants en fiscalité canadienne (12 numéros, offerts électroniquement), Actualités fiscales pour les propriétaires exploitants (4 numéros, offerts électroniquement), Canadian Tax Focus (4 numéros, offerts électroniquement) et le Rapport de la conférence annuelle. Les livres et monographies peuvent être achetés sur le site Web de la Fondation www.fcf-ctf.ca.
Revue Fiscale Canadienne — parution trimestrielle aux membres sur www.fcf-ctf.ca (Non-membres 75 $ par numéro, 343,75 $ par année). Bulletins Perspectives en fiscalité et en politique fiscal — parution trimestrielle disponible aux membres sur www.fcf-ctf.ca. Actualités fiscales pour les propriétaires exploitants — parution trimestrielle disponible aux membres sur www.fcf-ctf.ca. Canadian Tax Focus — parution trimestrielle disponible aux membres et non-membres sur www.fcf-ctf.ca. Faits saillants en fiscalité canadienne — archives (2000-2019) accessibles aux membres sur www.fcf-ctf.ca. Rapports des Conférences — comptes rendus des conférences annuelles sur la fiscalité (Membres 40 $; Non-membres 95 $). Dernière édition : 2018 (Membres 40 $; Non-membres 350 $). — Tax Dispute Resolution, Compliance, and Administration in Canada: Proceedings of the June 2012 Conference (Membres 30 $; Non-membres 195 $) — Collections contenant les travaux présentés aux conférences régionales sur la fiscalité, soit British Columbia, Prairie Provinces, Ontario et Atlantic Provinces, sont disponibles en format USB (Membres 445 $; Non-membres 495 $). Finances of the Nation — Analyse des recettes et dépenses, et quelques budgets, des gouvernements fédéral, provinciaux et locaux au Canada. Les copies PDF pour les années 2002-2012 sont disponibles sur le site Web de la FCF pour téléchargement gratuit. Dans le numéro 62:3 (2014), « Finances of the Nation » est apparu dans les éditions de la Revue fiscale canadienne à titre de nouvelle rubrique. Monographies 2019. Funding the Canadian City, Enid Slack, Lisa Philipps, Lindsay M. Tedds et Heather L. Evans, éds. ($40 chacun) 2018. Tax Treaties After the BEPS Project: A Tribute to Jacques Sasseville, Brian J. Arnold, éd. (Membres 60 $; Non-membres 90 $) 2018. Reforming the Corporate Tax in a Changing World, l’École de politique publique de l’Université de Calgary (Membres 30 $; Non-membres 50 $) 2017. Income Tax at 100 Years: Essays and Reflections on the Income War Tax Act, Jinyan Li, J. Scott Wilkie et Larry F. Chapman, éds. (Membres 60 $; Non-membres 90 $) 2016. Reform of the Personal Income Tax in Canada, l’École de politique publique de l’Université de Calgary (Membres 35 $; Non-membres 50 $) 2016. Canadian Taxation of Trusts, Elie S. Roth, Tim Youdan, Chris Anderson et Kim Brown (Membres 150 $; Non-membres 200 $; Étudiants 50 $) 2016. User Fees in Canada: A Municipal Design and Implementation Guide, Catherine Althaus et Lindsay M. Tedds (40 $ chacun) 2015. Timing and Income Taxation, 2 ième édition, Brian J. Arnold, Colin Campbell, Michael Hiltz, Richard Marcovitz, Shawn D. Porter et James R. Wilson (Membres 25 $; Non-membres 125 $; Étudiants 25 $)
■ xiv ■ 2015. Effective Writing for Tax Professionals, Kate Hawkins et Thomas E. McDonnell, QC (Membres 35 $; Non-membres 40 $) 2014. After Twenty Years: The Future of the Goods and Services Tax, l’École de politique publique de l’Université de Calgary (Membres 25 $; Non-membres 35 $) 2013. Essays on Tax Treaties: A Tribute to David A. Ward, Guglielmo Maisto, Angelo Nikolakakis et John M. Ulmer, éds. (100 $ chacun) 2012. Tax Policy in Canada, Heather Kerr, Ken McKenzie et Jack Mintz, éds. (Membres 75 $; Non-membres 100 $; Étudiants 50 $) 2011. Canadian Tax Foundation Style Guide, 5 ième édition (Membres 35 $; Non-membres 40 $) 2011. International Financial Reporting Standards: Their Adoption in Canada, Jason Doucet, Andrée Lavigne, Caroline Nadeau, Jocelyn Patenaude et Dave Santerre (Membres 30 $; Non-membres 40 $) 2011. Tax Expenditures: State of the Art — Selected Proceedings of the Osgoode 2009 Conference, Lisa Philipps, Neil Brooks et Jinyan Li, éds. (Membres 45 $; Non-membres 55 $; Étudiants 30 $) 2010. Taxation of Private Corporations and Their Shareholders, 4 ième édition (Membres 75 $; Non-membres 100 $; Étudiants 25 $) Collection Tax Professional (Prière d’indiquer le titre et le nom de l’auteur sur votre commande.) 2003. The Taxation of Business Profits Under Tax Treaties, Brian J. Arnold, Jacques Sasseville et Eric M. Zolt, éds. (édition brochée, 75 $) 2003. International Taxation in the Age of Electronic Commerce: A Comparative Study, Jinyan Li. Publié en collaboration avec l’Association fiscale internationale (chapitre canadien). (Membres 95 $; Non-membres 145 $; Étudiants 45 $) 1999. Contrer l’abus des conventions fiscales : Point de vue canadien sur une question internationale, Nathalie Goyette (75 $ chacun) Canadian Tax Paper Series (Prière d’indiquer le numéro de la publication.) No 112 : 2009. Des résponses efficaces aux planifications fiscales agressives : leçons à retenir des autres juridictions, Gilles N. Larin et Robert Duong, avec la contribution de Marie Jacques No 111 : 2009. Reforming Canada’s International Tax System: Toward Coherence and Simplicity, Brian J. Arnold (Membres 100 $; Non-membres 125 $) No 110 : 2006. Financing Education and Training in Canada, 2 ième édition, Douglas Auld et Harry Kitchen No 109 : 2004. The Canadian Federal-Provincial Equalization Regime: An Assessment, Alex S. MacNevin No 108 : 2004. Venture Capital and Tax Incentives: A Comparative Study of Canada and the United States, Daniel Sandler No 107 : 2002. Municipal Revenue and Expenditure Issues in Canada, Harry M. Kitchen (Membres 20 $; Non-membres 40 $) No 106 : 2002. Taxes and the Canadian Underground Economy, David E.A. Giles et Lindsay M. Tedds No 105 : 2000. The Income Tax Treatment of Financial Instruments: Theory and Practice, Tim Edgar No 104 : 1999. Rationality in Public Policy: Retrospect and Prospect, A Tribute to Douglas G. Hartle, Richard M. Bird, Michael J. Trebilcock et Thomas A. Wilson, éds. No 103 : 1999. Canadian Tax Policy, 3 : édition, Robin W. Boadway et Harry M. Kitchen No 101 : 1997. General Payroll Taxes: Economics, Politics, and Design, Jonathan R. Kesselman) No 102 : 1997. Financing the Canadian Federation, 1867 to 1995: Setting the Stage, David B. Perry No 100 : 1995. Growth of Government Spending in Alberta, Paul Boothe No 99 : 1995. Financing Education and Training in Canada, Harry Kitchen et Douglas Auld Special Studies in Taxation and Public Finance Series (Prière d’indiquer le numéro de la publication.) No 2 : 2000. Gambling and Governments in Canada, 1969-1998: How Much? Who Plays? What Payoff? François Vaillancourt et Alexandre Roy No 1 : 1998. Federal-Provincial Tax Sharing and Centralized Tax Collection in Canada, Ernest H. Smith
■ xv ■ ■ In the Research Centre
The Canadian Tax Foundation maintains for its staff and members a comprehensive refer - ence library of current and historical materials on taxation, public finance, and related subjects. The Douglas J. Sherbaniuk Research Centre contains one of the largest publicly accessible collections of tax information in the world. While the collection is built around a large base of Canadian materials, it also contains a significant portion of information covering international taxation, including country tax profiles, low-tax jurisdictions, inter - national tax treaties, and cross-border and international tax planning. The Research Centre currently holds domestic and international periodicals, case reporters, looseleaf services, and books. Housed at the Foundation’s Toronto headquarters, the Douglas J. Sherbaniuk Research Centre is open from 9 am to 5 pm, Monday to Friday. Staff are available for research assistance, particularly to members who cannot visit the Research Centre in person. Access is by e-mail (“Ask a Librarian” on our website or [email protected]), by telephone (416-599-0283, ext. 505), by facsimile (416-599-9283), or by toll-free (1-877-733-0283, ext. 505). The following books have been added to the Research Centre in recent months.*
CANADA Krishna, Vern. Fundamentals of Canadian Income Tax Volume 1: Personal Tax, 2d ed. Toronto: Thomson Reuters Canada, 2019. Sherman, David M., ed. Department of Finance Technical Notes: Income Tax, 31st ed. Toronto: Thomson Reuters Canada, 2019. Thomson Reuters Canada. Practitioner’s Income Tax Act, 56th ed., supplement. Toronto: Thomson Reuters Canada, 2019.
INTERNATIONAL Buijze, Renate, Tackling the International Tax Barriers to Cross-Border Charitable Giving: Philanthropy for the Arts in the Era of Globalization. IBFD Doctoral Series vol. 51. Amsterdam: IBFD, 2019. Debelva, Filip. International Double Taxation and the Right to Property: A Comparative, International and European Law Analysis. IBFD Doctoral Series vol. 20. Amsterdam: IBFD, 2019. International Fiscal Association. Congress Report IFA 2019: Summary of Proceedings of the 2019 London Congress. Rotterdam: IFA, 2019. Organisation for Economic Co-operation and Development. Country-by-Country Reporting— Compilation of Peer Review Reports (Phase 2). Paris: OECD, 2019. ______. Tax Administration 2019: Comparative Information on OECD and Other Advanced and Emerging Economies. Paris: OECD, 2019. Spies, Karoline. Permanent Establishments in Value Added Tax: The Role of Establishments in International B2B Trade in Services Under VAT/GST Law. European and International Tax Law and Policy Series vol. 13. Amsterdam: IBFD, 2019. Wheeler, Joanna, ed. The Aftermath of BEPS. Amsterdam: IBFD, 2019.
* Please contact the respective publishers if you wish to purchase any of the items listed.
■ xvi ■ canadian tax journal / revue fiscale canadienne (2020) 68:1, 1 - 3 1 https://doi.org/10.32721/ctj.2020.68.1.zeli
The Evaluation of Job Tax Incentives: An Analysis of a Regional Tax
Alessandro Zeli*
PRÉCIS On utilise souvent la déduction d’impôt à l’emploi comme outil de politique publique pour stimuler la croissance et la relance économique. L’analyse de l’incidence des dispositions de ce genre qui ont été adoptées dans un passé récent peut éclairer les effets des politiques fiscales actuelles. Cet article vise à étudier les effets d’une déduction fiscale de l’imposta regionale sulle attività produttive (impôt régional sur les activités productives), ou IRAP, accordée par l’Italie aux entreprises qui ont augmenté leur personnel entre 2005 et 2007. Les principaux objectifs de l’analyse sont d’évaluer l’augmentation et la permanence des nouveaux emplois, de discerner tout changement dans la structure de l’emploi des entreprises bénéficiaires, et d’évaluer l’efficacité des différents montants de déduction accordés aux entreprises des régions défavorisées afin de réduire l’écart d’emplois. Selon les résultats de l’analyse effectuée à l’aide d’un modèle de différence dans la différence, les entreprises qui ont bénéficié des incitations de l’IRAP ont enregistré des changements plus importants et plus durables des indicateurs sélectionnés par rapport aux entreprises n’appliquant pas la déduction, ce qui confirme l’efficacité de la disposition. La mesure adoptée prévoyait des déductions plus importantes pour les régions défavorisées du sud de l’Italie, mais les résultats n’indiquent pas une augmentation plus élevée de l’emploi dans ces régions.
ABSTRACT An employment tax deduction is frequently used as a public policy tool to stimulate economic growth and recovery. Analysis of the impact of such provisions adopted in the recent past may shed light on the effects of current tax policies. This article aims to estimate the effects of a tax deduction for Italy’s imposta regionale sulle attività produttive (regional tax on productive activities), or IRAP, granted to firms that increased their personnel between 2005 and 2007. The main objectives of the analysis are to assess the increase in, and the permanence of, new employment; to detect any changes in the
* Senior researcher, Istituto Nazionale di Statistica (Istat), Division for Data Analysis and Economic, Social, and Environmental Research (e-mail: [email protected]). I am grateful to Marco Ventura, whose many suggestions enhanced the quality of this article. I am also grateful to Antonella Caiumi, Lorenzo Di Biagio, and Marco Rinaldi for reading the article and providing useful comments. I bear sole responsibility for any errors or omissions.
1 2 n canadian tax journal / revue fiscale canadienne (2020) 68:1 employment structure of beneficiary firms; and to evaluate the effectiveness of different deduction amounts granted to firms in disadvantaged regions in order to reduce the employment gap. The results of the analysis using a difference-in-difference model indicate that firms enjoying IRAP incentives registered more significant and more enduring changes in the selected indicators as compared with firms not taking the deduction, thus verifying the effectiveness of the provision. The adopted measure provided for larger deductions for disadvantaged regions of southern Italy, but the results do not register a larger increase in employment in those regions. KEYWORDS: TAX DEDUCTIONS n ITALY n EMPLOYMENT POLICIES n EVALUATION n REGIONAL n TAX POLICY
CONTENTS Introduction 2 Prior Studies of the Impact of Tax Incentives on Employment 3 Aims of This Study 6 The IRAP Tax Deduction 9 The IRAP 9 The IRAP Deduction Introduced To Promote Employment 10 Data and Variables 12 Methodological Framework 13 DiD model 13 Robustness Check 18 Displacement Effects 19 Matched Sample 20 Difference-in-Difference Results 24 Results 24 Robustness Check 27 Displacement Effects 27 Conclusions 28 Appendix Efficiency Estimation 29
INTRODUCTION An employment tax deduction is frequently used as a public policy tool to stimulate economic growth and recovery. In many countries, tax deductions are implemented to help firms to hire new employees. For instance, Canadian political parties have included different job creation tax measures in their election platforms. Examples are a tax cut of 9-11 percent for small businesses (which are primary job creators) proposed in the federal New Democratic Party (NDP) platform for the 2011 general election, and a proposal for a tax credit to help businesses to invest in new job creation included in the Alberta NDP platform for the province’s 2015 election.1
1 New Democratic Party, Giving Your Family a Break: Practical First Steps, federal NDP platform, 2011 (www.documentcloud.org/documents/83745 -ndp- 2011 -platform.html); and Alberta New the evaluation of job tax incentives: an analysis of a regional tax n 3
On a regional basis, there are many other initiatives. For example, the Ontario co- operative education tax credit, which aims to create new fixed-term (four-month) employment contracts, mainly for students that are still enrolled in post-secondary programs, offers the employer a tax benefit of up to $3,000 for each newwork placement/contract; and the Ontario apprenticeship training tax credit encourages the hiring of trainees by providing a refundable tax credit based on salaries and wages paid to apprentices (capped at $5,000 annually for each apprenticeship, for a maximum of three years).2 The ex-post evaluation of these kinds of incentive provisions may shed light on their advantages and disadvantages, and contribute to the debate on their use. In Italy, a measure was enacted in 2005 providing a deduction for the imposta regionale sulle attività produttive (regional tax on productive activities), or IRAP, that was avail- able to Italian firms for new job creation. The deduction expired in 2008. In this article, the effects of this measure between 2005 and 2007 are analyzed. A frequently recurring question related to employment tax deductions is whether they lead to the creation of jobs that would not have been created in the absence of the incentive. In addressing this question, it is necessary to determine whether employment tax deductions generate additional effects relating to job creation and to identify any differential effects between regions. We contribute to this discussion by conducting a systematic investigation of how the Italian IRAP deduction affected employment and the benefiting firms’ performances linked to employment. The novelty of this research is the use of a database that permits the inclusion of firms of any size and with varying characteristics, in addition to the estimated impact of the deduction on employment. Our analysis describes in detail the impact of the IRAP deduction at the regional level, and how a broad set of variables (productivity, sales, and average salaries) responded to the provision, in order to gain an under- standing of the overall effect on firms’ employment policies. Finally, we compute the cost of obtaining such impact, broken down by industry and geographical zone, using information on the actual amount of the subsidy. In particular, we calculate the average cost of an additional job created as a result of the subsidy. These esti- mates permit a straightforward interpretation of the provision’s effects.
Prior Studies of the Impact of Tax Incentives on Employment A number of studies have been conducted to determine the impact of tax incentives on employment in both the United States and Europe.3 Indeed, in recent years,
Democratic Party, Leadership for What Matters: Election Platform 2015 (http://d3n8a8pro7vhmx .cloudfront.net/themes/5532a70aebad640927000001/attachments/original/1429634829/ Alberta_NDP_Platform_2015.pdf?1429634829). 2 See University of Waterloo, “Ontario Co-operative Education Tax Credit” (https://uwaterloo .ca/hire/funding-opportunities/tax-credits-and-incentives); and Ontario, Ministry of Finance, “Apprenticeship Training Tax Credit” (www.fin.gov.on.ca/en/credit/attc/index.html). 3 Jonathan R. Kesselman, Samuel H. Williamson, and Ernst R. Berndt, “Tax Credits for Employment Rather Than Investment” (1977) 67:3 American Economic Review 339 - 49; 4 n canadian tax journal / revue fiscale canadienne (2020) 68:1 many European governments have implemented tax incentives to encourage firms to expand employment. For EU countries, the European Commission encourages policies that enhance the flexibility of labour markets and promote increased labour market participation, such as reforms in tax benefit systems that may contribute to improving the functioning of EU labour markets.4 A statistical and econometric analysis by Carone et al.5 showed that a greater tax wedge has a significant negative impact on labour force participation and employment rates for non-EU member countries. More recent studies on this topic were carried out by Huttunen, Pirttilä, and Uusitalo6 and by Neumark and Grijalva.7 Both studies indicated positive effects of job subsidies and credits on employment using a difference-in-differenceD ( iD) methodology. Huttunen et al.8 analyzed the employment effects of a Finnish payroll- tax subsidy scheme targeted at older and low-wage workers. They found that there was an increase of approximately 1 percent in the employment rate of older workers as a result of the subsidy, but they did not find statistically significant effects for low- wage workers. They argued that the subsidy provision was inefficient owing to the high cost relative to the minimal effect on employment rates. Neumark and Grijalva9 analyzed state hiring credits in the United States. They did not find an effect on employment growth for the types of hiring credits that they examined. The results of their research indicated that a non-categorical and more broadly targeted hiring credit is substantially effective only during recessions. Moreover, evidence indicated
John H. Bishop and Mark Montgomery, Does the Targeted Jobs Tax Credit Create Jobs at Subsidized Firms? CAHRS Working Paper (Ithaca, NY: Cornell University, School of Industrial and Labor Relations, Center for Advanced Human Resource Studies, September 1991); Peter Bohm and Hans Lind, “Policy Evaluation Quality: A Quasi-Experimental Study of Regional Employment Subsidies in Sweden” (1993) 23:1 Regional Science and Urban Economics 51 - 65; Hildegunn E. Stokke, Regional Payroll Tax Cuts and Individual Wages: Heterogeneous Effects Across Education Groups, Working Paper series no. 16815 (Trondheim: Norwegian University of Science and Technology, Department of Economics, 2015); and Robert S. Chirinko and Daniel J. Wilson, Job Creation Tax Credits and Job Growth: Whether, When and Where? Working Paper 2010 - 25 (San Francisco: Federal Reserve Bank of San Francisco, 2010). 4 See Giuseppe Carone and Aino Salomäki, Reforms in Tax-Benefits System in Order To Increase Employment Incentives in the EU, Ecofin Economic Papers no. 160 (Brussels: European Commission, Directorate-General for Economic and Financial Affairs, September 2001). 5 Giuseppe Carone, Klara Stovicek, Fabiana Pierini, and Etienne Sail, Recent Reforms of the Tax and Benefit Systems in the Framework of Flexicurity, European Economy Occasional Papers (Brussels: European Commission, Directorate-General for Economic and Financial Affairs, February 2009) (https://doi.org/10.2765/92751). 6 Kristiina Huttunen, Jukka Pirttilä, and Roope Uusitalo, “The Employment Effects of Low-Wage Subsidies” (2013) 97:1 Journal of Public Economics 49 - 60. 7 David Neumark and Diego Grijalva, “The Employment Effects of State Hiring Credits” (2017) 70:5 Industrial and Labour Relations Review 1111 - 45. 8 Huttunen et al., supra note 6. 9 Neumark and Grijalva, supra note 7. the evaluation of job tax incentives: an analysis of a regional tax n 5 that the effects of state hiring credits are stronger when such measures target the unemployed. Another recent study10 analyzed a 2007 Swedish provision that substantially reduced the employer payroll tax for younger workers. The study’s findings showed a small impact on employment and wages; however, the effect on employment dif- fered across ages and over the business cycle. The provision was more effective for younger employees than for older workers; furthermore, the impact of the provi- sion appeared to be strongly procyclical. In the context of Italy, prior studies have considered IRAP provisions and deduc- tions aimed at employment growth. For example, Bordignon, Schmitz, and Turati11 focused on the relationship between reduction of the fiscal wedge (consisting of an IRAP deduction on the labour side) and the contemporaneous increase of value-added tax or fiscal devaluation. Their paper was based on a DiD model that analyzed the differential treatment of northern and southern Italian regions. The study revealed that the tax cut significantly increased the employment rate and that doubling the deduction increased the employment rate by about 2 percent. Many subsidies and grants may be differentiated at the regional level or may be provided directly by local authorities. In such cases, the differential impact of a pro- vision at the regional level can be studied. In the United States, the findings of studies administered locally in several states12 suggested evidence of job creation in response to tax incentives, specifically employment tax credits. Gabe and Kraybill13 analyzed the effects of economic development incentives on employment growth at the state level in the United States between 1993 and 1995, while Porro and Salis14 recently analyzed four types of incentives aimed at firm growth provided in the Lombardy region in northern Italy. Neither analysis revealed evidence that those provisions had a positive impact on employment growth at the regional level. On the other hand, Shuai and Chmura15 carried out a state-level study in the United
10 Niklas Kaunitz and Johan Egebark, Payroll Taxes and Firm Performance, IFN Working Paper no. 1175 (Stockholm: Research Institute of Industrial Economics, 2017) (revised April 8, 2018). 11 Massimo Bordignon, Marie-Luise Schmitz, and Gilberto Turati, “Does Fiscal Devaluation Really Work? Evidence from an Italian Experiment,” paper presented at CESifo Area Conference on Public Sector Economics, April 16 - 18, 2015. 12 Dagney Faulk, “Do State Economic Development Incentives Create Jobs? An Analysis of State Employment Tax Credits” (2002) 55:2 National Tax Journal 263 - 80 (https://doi.org/10.17310/ ntj.2002.2.04); and Jungyul Sohn and Gerrit-Jan Knaap, “Does the Job Creation Tax Credit Program in Maryland Help Concentrate Employment Growth?” (2005) 19:4 Economic Development Quarterly 313 - 26. 13 Todd M. Gabe and David S. Kraybill, “The Effect of State Economic Development Incentives on Employment Growth of Establishments” (2002) 42:4 Journal of Regional Science 703 - 30. 14 Giuseppe Porro and Valentina Salis, “Do Local Subsidies to Firms Create Jobs? Counterfactual Evaluation of an Italian Regional Experience” (2018) 97:4 Papers in Regional Sciences 1039 - 56. 15 Xiaobing Shuai and Christine Chmura, “The Effect of State Corporate Income Tax Rate Cuts on Job Creation” (2013) 48:3 Business Economics 183 - 93. 6 n canadian tax journal / revue fiscale canadienne (2020) 68:1
States to analyze the impact of a tax cut on employment growth. They found a posi- tive relationship between the tax cut and growth of employment for those states that cut their corporate taxes, and they verified a decrease in the gap between states with higher employment growth and states with lower employment growth. Another study based on US state-level data16 found that any increase in state tax on dividend income reduced tax revenue, while states benefited, in terms of revenue, from an increase in tax on income from wages or in sales or property tax. Similar incentives have been implemented in Canada in the past. For instance, the Cape Breton investment tax credit aimed to spur investment, employment, and productivity in Cape Breton, although the effectiveness of this provision was questioned.17 In the context of France, a study that aimed to determine the most efficient instrument to fill the gap in economic performance (productivity) at the regional level18 found that providing a financial incentive and granting an investment credit were less costly policies than the alternative option of lowering the corporate tax rate. Finally, a UK study by Devereux, Griffith, and Simpson19 aimed to analyze how the impact of government grants influenced firms’ decisions on the location of their plants. The results showed that the location decisions of firms were influenced more by the local industry structure than by government grants, which were found to be of little importance.
Aims of This Study In Italy, some incentives in the form of deductions from the IRAP tax base were introduced in the last decade to promote employment. These measures included incentives for small firms, facilitation of job placement for disadvantaged employ- ees, incentives for transitioning from fixed-term employment contracts to permanent (open-ended) employment contracts, and help for disadvantaged regions. Owing to the peculiar characteristics of the IRAP, a large portion of employment tax deduc- tions in Italy are implemented through IRAP deductions.20
16 Hakan Yilmazkuday, “Individual Tax Rates and Regional Tax Revenues: A Cross-State Analysis” (2017) 51:5 Regional Studies 701 - 11. 17 Michael Daly, Ian Gorman, Gordon Lenjosek, Alex MacNevin, and Wannakan Phiriyapreunt, “The Impact of Regional Investment Incentives on Employment and Productivity: Some Canadian Evidence” (1993) 23:4 Regional Science and Urban Economics 559 - 75. 18 Hélène Laurent, Michel Mignolet, and Olivier Meunier, “Regional Policy: What Is the Most Efficient Instrument?” (2009) 88:3Papers in Regional Science 491 - 507. 19 Michel P. Devereux, Rachel Griffith, and Helen D. Simpson, “Firm Location Decisions, Regional Grants and Agglomeration Externalities” (2007) 91:3 - 4 Journal of Public Economics 413 - 35. 20 This provision did not overlap any other subsidy granted to firms before or after the period considered here. The actual sample data span the years 2004 - 2010, which include one year before the introduction of the provision and three years after the last deduction was obtained. the evaluation of job tax incentives: an analysis of a regional tax n 7
Our contribution is to estimate the impact of the IRAP tax deduction on employ- ment using a large sample of Italian firms and applying a reduced-form approach to empirical modelling. We explore the effects of the IRAP deduction on employment growth from the following perspectives:
n the durability of the increasing effects of the tax deduction on employment owing to a change in the occupational structure of firms; n the reduction of the employment gap between more developed regions and regions that are disadvantaged; n increasing employment of a skilled workforce; and n an increase in productivity in the form of an increase in market share, as indicated by increased sales.
Other studies have examined similar measures and similar incentives related to the IRAP. For instance, Monteduro et al.21 investigated an IRAP tax credit granted for job hiring after 2007; however, this study focused only on small and medium- sized enterprises (SMEs) located in southern regions, and it did not calculate the average cost of an additional job. Recently, Porro and Salis22 considered four pro- grams granted at the local level by the Lombardy region that provided financial incentives for job-hiring firms, with a partial focus on territory regarding the targets of the provisions. However, the authors implemented only a short-run analysis, so some questions remain about the long-run effects. Kangasharju23 analyzed the wage subsidy policy in Finland. This provision is quite different from a tax credit, since the subsidies aim to increase wages up to the minimum level granted in Finland. Hence, “wage subsidies are directed to firms who employ the kind of unemployed whose productivity and qualifications are lower than the levels needed in active labour markets.”24 In other words, the instru- ment is not targeted to a generalized public. As stated above, the IRAP deduction generated some questions. For example, is there a real differential in employment growth as between beneficiaries and non- beneficiaries? Is there a greater employment change for southern regions given the greater deductions granted for firms located there? Are the effects of the deduction on employment permanent? Does the deduction affect the performance of non- beneficiary firms? How efficient is the provision in terms of cost per new job created
21 Maria Teresa Monteduro, Marco Manzio, Federica Alivernini, Daniela Bucci, Jacopo Canello, Febrizio De Grandis, Marco Mastracci, and Paolo Pavone, “Employment Tax Credits and Job Creation Trends in SMEs: Evidence from a Policy Intervention in Italy,” paper presented at the 2014 Counterfactual Methods for Policy Impact Evaluation (COMPIE) Conference, Rome, 2014. 22 Porro and Salis, supra note 14. 23 Aki Kangasharju, “Do Wage Subsidies Increase Employment in Subsidized Firms?” (2007) 74:293 Economica 51 - 67. 24 Ibid., at 57. 8 n canadian tax journal / revue fiscale canadienne (2020) 68:1 in different regions? To address these questions, we carried out a two-step proced- ure (using propensity score matching and DiD), which identified a firm as “treated” if it benefited from the IRAP deduction after hiring an employee with a permanent contract (subject to the rules and limitations discussed in the next section). This implies greater deductions from the IRAP base with respect to non-beneficiary firms. To sum up, the aim of this article is multifold. We try to verify not only the em- ployment increase for beneficiaries, but also whether the change was enduring. In other words, we want to verify the existence of a differential effect on another dimension—the duration of the employment increase. The methodology used in estimating differential effects can help to determine whether the differential change in employment level is permanent (an important consideration), and whether the differential effects calculated for other variables, such as productivity, salary, and sales, are actual as well. Many authors studying economic measures have utilized, as an outcome variable, the target of the provisions. Kangasharju25 analyzed the effects of a wage subsidy for new workers on employment and employees’ skill and wage level by means of a DiD methodology. Other studies analyzed the effects of employment tax deductions on total employment growth for all firms located in southern Italian regions, by means of “average treatment effect on the treated” AT( T) estimates.26 In the same way, analyses of the impact of tax policies on other economic variables, carried out by means of DiD methodologies, have considered the provision target as an outcome variable. For instance, Haegeland and Møen27 and Bozio, Irac, and Py28 examined the effects of a tax deduction for investment in research and development R & D( ) on the increase in such investments, while Bronzini and de Blasio29 investigated the effect on investment growth of investment incentives granted by Law 488/12 in Italy. Our study analyzes an extensive and generalized provision not targeted to specific firms or employees, but aimed at improving employment all over the country. It represents a comprehensive analysis of this particular measure—a tax deduction granted on the basis of a regional tax—implemented in 2005-2007, which also compares the effects at the regional level. The rest of the article is organized as follows. The next (second) section describes the IRAP tax deduction. The third section describes the data sources and summary
25 Kangasharju, supra note 23. 26 See, for example, Monteduro et al., supra note 21. 27 Torbjørn Haegeland and Jarle Møen, Input Additionality in the Norwegian R & D Tax Credit Scheme, Report no. 2007/47 (Oslo: Statistics Norway, December 2007). 28 Antoine Bozio, Delphine Irac, and Loriane Py, Impact of Research Tax Credit on R & D and Innovation: Evidence from the 2008 French Reform, Banque de France Working Paper no. 532 (Paris: Banque de France, December 2014). 29 Raffaello Bronzini and Guido de Blasio, “Evaluating the Impact of Investment Incentives: The Case of Italy’s Law 488/1992” (2006) 60:2 Journal of Urban Economics 327 - 49. the evaluation of job tax incentives: an analysis of a regional tax n 9 statistics for this study, and the fourth section outlines the study methodology. The fifth section explains the evaluation of the results of the matching procedure, and the sixth section presents the results of the DiD estimation for employment and other indicators. The conclusions are reported in the last section. An appendix to the article provides an estimate of the efficiency of theIRAP tax deduction.
THE IRAP TAX DEDUCTION The IRAP The IRAP is a regional production tax adopted in 1998 as part of a reform package designed to remedy certain weaknesses in the Italian tax system. The IRAP repre- sents perhaps the most important modern move toward a local business value-added tax. It is similar, in some respects, to the French territorial economic contribution (contribution économique territoriale); for instance, both programs use a method of allocating revenues among localities that is essentially based on local employ- ment, as described in Gilbert.30 Implementation of the IRAP was intended to address multiple issues, including avoiding the imposition of high statutory tax rates on income or additional taxation on business net worth, and to provide regions with an autonomous source of revenue to increase their fiscal accountability, specifically in the health-care sector. The tax burden was no longer focused on gross profits, but instead was based on a broader definition of corporate revenues to allow for a re- duction in rates. The IRAP’s tax base is the net value added generated by all types of businesses in each region. Thus, the tax is based not on net income before tax, but on the value added produced by a company at the local level. Value added is defined as the dif- ference between operating revenues and costs before interest income and expenses, and, in particular, labour costs are deducted, so that part of the tax affects values other than the company’s pure income. As well as differentiating tax base from gross profit, theIRAP is also payable in the event of a reported loss. This provision makes the tax procyclical and thus may worsen the already difficult situation of companies during economic and financial crises. Given the broad tax base, the ordinary statutory rate applied at the time the IRAP was enacted was relatively low—though, notwithstanding the low rate, the tax rev- enue was considerable—and the initial tax rate has been reduced in the last decade. As for the definition of the tax base, specific rates apply to different sectors of -eco nomic activity and different organizational forms. If the activity is conducted in several regions, the value is divided according to the remuneration payable to staff in each region. Once the tax has been collected, the revenues are allocated among regions in proportion to labour costs incurred in each region, net of a share to be paid to the Italian state, as compensation for costs associated with control and cer- tification activities. The regions then distribute a share of the income toeach
30 Guy Gilbert, “Finances Publiques Locales Enjeux et Perspectives,” paper presented at the Institut National des Études Territoriales, Strasbourg, France, May 31 -June 1, 2010. 10 n canadian tax journal / revue fiscale canadienne (2020) 68:1 province and municipality. The introduction of the IRAP, moving taxation toward the regional level, implies a better balance between national and subnational tax systems. Almost all IRAP revenue is allocated to health-care expenditures, the main field of competence of the regions. Since 2005, regions can modify the tax rate by up to 1 percent when running into deficits with health-care expenditures. This option became compulsory in 2006 and since then has resulted in increased IRAP rates in the Abruzzo, Campania, Lazio, Molise, and Sicilia regions. Since the implementation of the IRAP, various incentives (in the form of deduc- tions from the tax base) have been introduced to promote small businesses and permanent employment, and to reduce regional gaps. The IRAP remains distinct from the corporate tax (IRES), even if it is possible to deduct some IRAP items from general corporate tax liabilities. (For an in-depth analysis of the IRAP, see Bor dignon et al.)31
The IRAP Deduction Introduced To Promote Employment Law n. 80 (2005) and Law n. 311 (2004) (Budget Law for 2005) introduced an important IRAP deduction aimed at increasing the employment base. The deduc- tion, operative from 2005 until 2008, when it expired,32 was intended to promote employment through permanent contracts and to improve employment rates to a greater extent in disadvantaged regions relative to the rest of the country. The provision covered both full-time and part-time permanent contracts, but was condi- tional on an actual increase in a firm’s employment base. The structure of theIRAP deduction influenced the effectiveness of the provision in creating jobs. The most relevant features of the deduction include the following:
n The IRAP’s new employment deduction was available for the creation of new, permanent contract jobs, so the deduction was not available for fixed-term contract jobs. n The deduction applied against the IRAP tax base; firms with no IRAP tax lia- bilities could not use the deduction. n The provision allowed for the deduction of labour costs up to € 20,000 for each new employee if the average number of permanent employees, calcu- lated on an annual basis, increased relative to the previous year. n An increase in the level of employment was realized if the difference between the number of employees at the end of a year and the annual average of the number of employees in the previous year was positive. n The deduction was maintained if the total employment did not increase in the fiscal years following the fiscal year for which the deduction was granted.
31 Bordignon et al., supra note 11. 32 In this study, we considered the deduction effect only for the three-year period before the financial crisis (that is, 2005 - 2007). the evaluation of job tax incentives: an analysis of a regional tax n 11
n The deduction aimed to provide additional permanent employment and was not available for replacement workers (employees replacing laid-off workers). n The amount of the deduction granted varied, depending on the location of the firm. This provision was intended to foster employment in less-developed regions; thus, a greater deduction was allowed if a firm was located in one of the less-developed regions in southern Italy (Campania, Puglia, Basilicata, Calabria, Sicilia, Sardinia, Abruzzo, or Molise).33 n A firm had to maintain the additional jobs created until the end of the 2008 financial year in order to maintain the employment base increase. n The deduction was extended to all types of industries. n A firm could obtain the IRAP deduction for one or more years. For example, a firm could obtain the deduction for 2005 and 2006, or a beneficiary firm in 2006 could obtain a further deduction (for other new employees) in 2007. n Finally, the deduction granted for each new job in a disadvantaged area (a southern region) increased during the period 2005 - 2007.
These provisions can be summarized as follows. Beneficiary firms had to increase their employment base from the average of the preceding financial year before receiving the deduction. It was possible for the same firm to decrease its employ- ment base in subsequent financial years, but only if the employment change remained positive compared to the initial year until financial year 2008. The bene- fit for the firm was achieved by means of deductions from the IRAP tax base, which includes the cost of labour. An example of the calculation of the number of deduc- tions granted to a hypothetical firm is presented in table 1. In order to benefit from the provision, eligible enterprises must be informed about the existence of the incentive and apply if the expected benefits exceed the expected cost of application. The costs incurred by firms in becoming beneficiaries of the IRAP deduction fall into several categories: search costs, compliance costs, reporting costs (linked to providing additional information to fiscal authorities), and hiring costs. Depending on those costs, firms in different industries and of different sizes may find it more or less advantageous to exploit tax deductions linked to new employment; moreover, they must maintain a positive employment differential relative to the initial year. Access to the deductions differed according to firms’ characteristics. In 2005, more “younger” firms, in business for fewer years, accessed the deduction than “older” firms. Newer firms are more dynamic and better able to accommodate the administrative costs referred to above. For the same reasons, the larger (and best- organized) firms accessed the deduction more often than smaller firms. The geographical and economic-sector classifications also mattered: in the northern
33 The additional deduction differs from one region to another. The calculation of the deduction is more complicated in these cases, but it can be estimated at around three to five times the base deduction granted for the rest of the country. 12 n canadian tax journal / revue fiscale canadienne (2020) 68:1
TABLE 1 Calculation of Deductions Granted to a Hypothetical Firm
2005 2006
Number of employees at Number of employees at 31/12/2004 ...... 22 31/12/2005 ...... 27 Hirings in 2005 ...... 5 Deductions granted for 2005 . . . 5 Number of employees at Number of employees at 31/12/2005 ...... 27 31/12/2006 ...... 25 Average number of employees in 2004 ...... 26.3 Dismissed in 2006 ...... 5 Employment growth, 2005 . . . . 0.7 Hirings in 2006 ...... 3 Average number of employees Deductions granted for 2005 . . . 5 in 2005 ...... 26.5 Employment growth, 2005 . . . . −1.5 Deductions granted for 2006 . . . 5 Granted for 2005 ...... 5 Granted for 2006 ...... 0
Note: Employment growth is calculated as the difference between the prior-year average and the current-year number of employees at year-end. Assumed financial year-end is December 31. Source: Agenzia delle Entrate (Italian Revenue Agency), Administrative Circular, February 13, 2006, n. 7/E. regions, in 2005, the share of beneficiaries was higher than in the southern regions; and industries (manufacturing and energy supply) received a larger share of the benefits than the service sector. These facts may explain why some eligible firms were not beneficiaries. They also contribute to the construction of the profile of the beneficiary firm used in the matching procedure described below.
DATA AND VARIABLES We exploited two main sources of data: corporate tax return data from the Italian Ministry of Economy, including all tax forms related to Italian limited enterprises (approximately 1,050,000 firms per year); and the balance sheet data of limited enterprises, a database acquired by the Italian Institute of Statistics (Istat) from the Chamber of Commerce. We identified eligible firms as those described in a study by Caiumi,34 or firms for which the IRAP economic base was greater than zero. We assembled a panel of approximately 1.2 million observations relating to about 168,000 firms per year over the seven years from 2004 through 2010.35 We considered only those firms
34 Antonella Caiumi, The Evaluation of the Effectiveness of Tax Expenditures—A Novel Approach: An Application to the Regional Tax Incentives for Business Investment in Italy, OECD Taxation Working Paper no. 5 (Paris: Organisation for Economic Co-operation and Development, 2011). 35 We considered a span of time longer than 2005 - 2007 in order to have data for a one-year period before the introduction of the provision and a three-year period after the last deduction was received. the evaluation of job tax incentives: an analysis of a regional tax n 13 that remained in the panel for at least five consecutive years and that received a certain deduction amount; hence, having at maximum only seven presences, we obtained a quasi-balanced panel with very little attrition, concentrated in the first year (2004) and in the last two years (2009 and 2010). This panel is representative of incorporated enterprises. In this study, the following variables were considered:
n ROA (return on assets): calculated as operating income (OI) over total assets; n Tangibles: defined as the ratio of tangible fixed assets to total assets; n Age: information about a firm’s age, in years; n Tax status: if the firm had corporate tax IRES( ) liabilities, tax status = 1; other- wise, tax status = 0; n Persons employed: the number of persons employed at the firm; n Sales: annual turnover; n Average salary: salary amount per head; n Productivity: the ratio of value added to persons employed; and n Export tendency (Exp_tend): export amount on turnover.
Firms were classified according to some categorical variables. Firm size was div- ided into three large revenue (turnover) classes: small (under € 2 million), medium (from € 2 million to € 10 million), and large (over € 10 million). Two other classifica- tions were considered: economic sector classification and geographical area classifi- cation. The use of a geographical area classification allowed for the disentanglement of the differential area deduction effects in the probit model described below. A summary of the sample firms’ statistics for the variables considered in this study is presented in table 2. From 2005 to 2007, the specific deduction amount for permanent contracts tripled for the firms included in the panel, while the number of beneficiaries -in creased by 20 percent. Consequently, the average firm deduction doubled in the period for the sample firms (table 3). In general, the deduction amount increased with firm size, and the distribution of size categories remained substantially unaltered during the period. Southern regions received increasing average deductions owing to the additional deduction granted to them at the local level.
METHODOLOGICAL FRAMEWORK DiD model This empirical study aims to determine whether exploiting the IRAP tax deduction had a causal effect on firms’ economic indicators, by reference to a targeted eligible firm. Following the approach and notation used by Blundell and Costa Dias36 and
36 Richard Blundell and Monica Costa Dias, “Evaluation Methods for Non-Experimental Data” (2000) 21:4 Fiscal Studies 427 - 68. 14 n canadian tax journal / revue fiscale canadienne (2020) 68:1
TABLE 2 Summary Statistics, 2004-2010
Standard Variable Mean deviation Median
Persons employed ...... 26.8 240.9 9.0 Average salarya ...... 23.2 19.2 21.1 ROA ...... 5.8 10.5 4.8 Tax status ...... 0.8 0.4 1.0 Age ...... 17.2 12.3 14.0 Productivitya ...... 47.3 91.4 36.2 Salesa ...... 6,608 6,610 1,268 Export tendency ...... 0.04 0.15 0.02 Tangibles ...... 0.21 0.22 0.1 a Thousands of euros.
TABLE 3 Annual IRAP Deductions, Sample Firms, 2005-2007
2005 2006 2007
Total IRAP deductiona ...... 1,549,865 3,203,217 4,544,177 Number of beneficiaries ...... 71,697 80,893 86,200 Average firm deductiona All firms ...... 22 40 53 By geographic area North-west ...... 21 38 46 North-east ...... 21 32 45 Centre ...... 21 35 48 South ...... 24 57 79 By firm size (turnover classes) Less than € 2 million ...... 12 19 29 € 2 million to € 10 million ...... 24 43 60 Over € 10 million ...... 85 163 181
IRAP = regional tax on productive activities. a Thousands of euros. by Bandick and Karpaty,37 we adopted a two-stage strategy.38 We first constructed a sample of matched beneficiary and non-beneficiary firms, and then we estimated a DiD coefficient for the matched sample. For this purpose, we developed the following formula and series of equations. Let TC ∈{0,1} be an indicator of whether firm i is a beneficiary of the IRAP deduc- 1 tion in time period t, and let y i, t + s be employment at time t + s; s > 0, after the first
37 Roger Bandick and Patrik Karpaty, “Employment Effects of Foreign Acquisition” (2011) 20:2 International Review of Economics and Finance 211 - 24. 38 Also described in Alessandro Zeli, “The Impact of ACE on Investment: The Italian Case” (2018) 35:3 Economia Politica 741 - 62. the evaluation of job tax incentives: an analysis of a regional tax n 15
0 deduction year. If firm i is a non-beneficiary firm, its outcome is denoted as y i, t + s . The causal effect on employment by being an IRAP tax deduction beneficiary for firm i at time t can be defined as
1 0 y i, t + s - y i, t + s . (1)
1 0 Now it is possible to observe y i, t + s , while y i, t + s is not observable; this is the pri- mary problem in the estimation of causal effects. So it is possible to define the average effect of exploiting the tax deduction as
1 0 1 0 E{ y i, t + s - y i, t + s |TCit = 1} = E{ y i, t + s |TCit = 1} - E{ y i, t + s |TCit = 1}. (2)
The last term of equation 2 is the counterfactual, which is difficult to construct. In other words, we must estimate what the outcome for firms exploiting the tax deduction would have been, on average, had those firms not used the tax deduction. One method to determine this estimation is to use the average employment of firms 0 that were not deduction beneficiaries, expressed as E{ y i, t + s |TCit = 0}. Variable TCit is influenced by many other variables and simultaneous effects being endogenously so determined. If these influences are not considered, the estimation of causal effects will be biased. Since this is a missing-data problem, we need to use the available information to impute the relevant information that is not possible to observe. However, if the selection into treatment is completely determined by a set of exogenous covariates (X), and if those covariates can be observed by the researcher and the assignment into treatment is random, then the outcomes for the non-treated firms are independ- ent of the participation status. Our approach employed a matching technique. Matching involves pairing bene- ficiary with non-beneficiary firms on the basis of similar pre-provision characteris- tics (X), including size, location, age, and profitability. Using this technique, we built a sample of non-recipient firms twinned with recipient firms to better approxi- mate the non-observed counterfactual event in equation 2. We used the Rubin39 and Rosenbaum and Rubin40 propensity score matching methodology. Under the con- ditional independence assumption, the selection occurs only on observables. A probit model was used to estimate the probability (or propensity score) of being a beneficiary firm. This is the first step in implementing propensity score matching. We begin with the following formula:
p(TCit = 1) = F (Xit - 1,Dj,Dt), (3)
39 Donald B. Rubin, “Assignment to Treatment Group on the Basis of a Covariate” (1977) 2:1 Journal of Educational Statistics 1 - 26. 40 Paul R. Rosenbaum and Donald B. Rubin, “The Central Role of the Propensity Score in Observational Studies for Causal Effects” (1983) 70:1Biometrika 41 - 55. 16 n canadian tax journal / revue fiscale canadienne (2020) 68:1
where TCit = 1 denotes a non-beneficiary firm in year t − 1 that benefits from tax deduction in year t; Xit − 1 is a vector of relevant firm-specific variables in year t − 1 that may influence the firm’s probability of being a beneficiary in year t; Dj controls for industry or area effects; andD t controls for time-fixed effects. The following equation explicitly indicates the variables included in the probit model.
2 p(TCit = 1) = b0 + b1Aget - 1 + b2Age t - 1 + b3TSt - 1 + b4ROAt - 1 + b5Exp_tendt - 1 + b6Tangiblest - 1 + d1Size + d2 Area + d3Industry + e. (4)
Using the propensity scores after the probit model estimation, it is possible to select the control firms for which the propensity score determines the closest match with a treated firm. We utilized the Stata procedure PSMATCH2 developed by Leuven and Sianesi41 to match treated and control firms. To identify the counter- factual, the main estimation method adopted was nearest-neighbour matching without replacement. When the control group of firms is identified, the DiD estimator can be used to estimate the impact of use of the tax deduction on the firms’ economic indicators. According to Wooldridge,42 this estimate can be obtained by employing the follow- ing regression:
yit = b0 + b1TCAi + b2 Afterit + s + b3TCAi × Afterit + s + b4Xit + εit, (5) where yit is the target outcome variable (change in employment). TC is a dummy vari- able equal to 1 for beneficiary (treated) firms and equal to 0 for non-beneficiary firms C. We multiply this dummy by the total deduction amount 43 for the single ith firm over the period A( i) to exploit this micro-level information, and thus obtain the 44 variable TCA as shown in Angrist and Pischke. The dummy variable Aftert + s takes a value of 1 in the post-tax deduction year t + s and a value of 0 in the year before the tax deduction. This dummy variable captures aggregate period effects that are common between the two groups T and C. The last term TCAi × Aftert + s represents the interaction between TCAi and Aftert + s . The coefficient of this last term (β3)
41 Edwin Leuven and Barbara Sianesi, “PSMATCH2: Stata Module To Perform Full Mahalanobis and Propensity Score Matching, Common Support Graphing, and Covariate Imbalance Testing,” Statistical Software Components, Boston College Department of Economics, 2003 (http://ideas.repec.org/c/boc/bocode/s432001.html). 42 Jeffrey M. Wooldridge,Econometric Analysis of Cross-Section and Panel Data (Cambridge, MA: MIT Press, 2002). 43 In thousands of euros. 44 Joshua A. Angrist and Jörn-Steffen Pischke,Mostly Harmless Econometrics: An Empiricist’s Companion (Princeton, NJ: Princeton University Press, 2009). the evaluation of job tax incentives: an analysis of a regional tax n 17 represents the DiD estimator of the effect of being a beneficiary on treated firm T, or β3 = γt + s . The DiD estimator eliminates unobserved time-invariant differences in employment between beneficiary and non-beneficiary firms. Finally,X t + s is a set of control dummy variables: geographical area, size, industry, and year. Moreover, as in Bandick and Karpaty,45 we included a vector of firm characteristics to control for differences in observable attributes among firms. In particular, following the literature on similar topics,46 sales and a measure of profit RO( A) were included in the control variables matrix X. Regression 4 was replicated for each of the variables of interest (turnover, average salary, and productivity). An important hypothesis underlying the DiD procedure is the parallelism as- sumption—that, without the deduction, the supported firms would have followed a trajectory parallel to that of the non-supported firms. However, there is noth- ing to support the truth of that assumption, as discussed in a 2012 report by the Associazione per lo Sviluppo della Valutazione e l’Analisi delle Politiche Pubbliche (ASVAPP).47 To address this issue, the employment trends for beneficiaries and non-beneficiaries in the matched sample and the probably natural dynamic for beneficiaries (that is, employment without the IRAP deduction) were further analyzed. The results are presented in figure 1. The beneficiaries and non-beneficiaries have parallel trajec- tories before 2005 (the starting year for IRAP deductions). After 2005, there is an increase in employment for beneficiaries that surpasses the increase for non- beneficiaries. The beneficiaries’ trajectories remain divergent from those ofthe non-beneficiaries throughout the three-year period 2005 - 2007. After 2008, the trajectories resume a parallel path, showing an impact of the tax deduction provision and a constant differential between beneficiaries and non- beneficiaries owing to the provision itself. The graph does not show effects in the years before the provision was adopted. The beneficiaries’ employment impact trend displays a strong increase in the first years after the provision was enacted (2005 - 2007); subsequently, the trend flattens out, indicating a permanent higher employment level for the beneficiaries.48 Hence, figure 1 supplies evidence of a common underlying trend (the natural dynamic) for treated and control firms, and a treatment effect producing a deviation from that trend.
45 Bandick and Karpaty, supra note 37. 46 Bronzini and de Blasio, supra note 29; and Kangasharju, supra note 23. 47 Associazione per lo Sviluppo della Valutazione e l’Analisi delle Politiche Pubbliche (ASVAPP), Counterfactual Impact Evaluation of Cohesion Policy: Impact and Cost-Effectiveness of Investment Subsidies in Italy, final report to the European Commission (DG Regional Policy) (Torino: ASVAPP, June 2012). 48 David H. Autor, “Outsourcing at Will: The Contribution of Unjust Dismissal Doctrine to the Growth of Employment Outsourcing” (2003) 21:1 Journal of Labor Economics 1 - 42, at 25-26; and Angrist and Pischke, supra note 44, at 178-80. 18 n canadian tax journal / revue fiscale canadienne (2020) 68:1
FIGURE 1 Decomposing Observed Change into Impact and Natural Dynamic Effects—Employment for the Matched Sample 900 800 700 Impact 600 ’000s) 500 Period for which 400 the deduction was granted 300 Employment ( 200 100 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Non-beneficiary Beneficiary Natural dynamic
Note: The employment natural dynamic is calculated for beneficiary firms by multiplying the beneficiaries’ employment level at year t − 1 by the employment annual increments registered for non-beneficiaries.
Robustness Check To assess the robustness of our results, we used an alternative format in which pro- pensity score matching and DiD are computed simultaneously, as shown by Blundell and Costa Dias.49 In this case, the DiD matching estimator is given by the equation
1 ATT = Σ __ [ y T - y C - Σ ijw ( y C - y C )], (6) i∈T ∩S N T i, t + s i, t - 1 j∈C∩S ij j, t + s j, t - 1 where ATT is the average effect of treatment on treated; N T is the number of bene- T ficiary firms; y i, t + s is the outcome variable (employment) for beneficiary firm i at C C year t + s; y j, t + s and y j, t - 1 are the outcome variables for non-beneficiary firm j at years t + s and t − 1, respectively; S denotes the region of common support; and the weights wj are defined by wj = ∑iwij. We use a nearest-neighbour matching al- gorithm, for which standard errors are obtained by bootstrapping, as described in Becker and Ichino.50
49 Richard Blundell and Monica Costa Dias, “Alternative Approaches to Evaluation in Empirical Microeconomics” (2009) 44:3 Journal of Human Resources 565 - 640. 50 Sascha O. Becker and Andrea Ichino, “Estimation of Average Treatment Effects Based on Propensity Scores” (2002) 2:4 Stata Journal 358 - 77. the evaluation of job tax incentives: an analysis of a regional tax n 19
Displacement Effects In our analysis, we assumed that the potential outcome for firms located in one re- gion did not depend on whether firms located in another region (or in the same region) received treatment or not. The identification of the treatment effect was based on the premise that subsidized firms did not influence the actions of non- subsidized or less-subsidized firms. Among the factors that could influence non-treated firms, given the presence of treated ones, were the competitive edge of treated firms owing to the decrease in the marginal cost of labour and the reallocation of spared funds for labour costs to marketing intensification. A negative effect may be the gain in market share for beneficiaries relative to non-beneficiaries. A positive effect of treatment on non- treated firms is the tendency for those firms to have increasing employment because of favourable market prospects and an improved business climate. Moreover, if there was different treatment at the regional (or geographical zone) level, firms located in less-subsidized zones may be inclined to move to more-subsidized zones. This effect may occur among different geographical areas. Such movement would imply an effect on a positive net inflow of firms because of the incentive to relocate to areas with more generous deduction schemes, as analyzed by Bennmarker, Mellander, and Öckert51 and by Bordignon et al.52 If these factors were not negli- gible, the estimate included both the effect of treatment on the treated firms and the effect of subsidies on other firms. Hence, the counterfactual employment estimate decreased/increased (since the employment of non-subsidized firms decreased/ increased), and estimates were biased upward/downward. We followed the approach of Kangasharju53 to address the first displacement effect. The displacement effects most likely occur within the same industry and/or geographical area. There are three potential displacement effects, generated by firms within the same industrial sector, within the same region, or within the same sector and region as the beneficiary firms. In building the industry-specific displace- ment variable, it was necessary to sum the industry-specific deduction amounts for each ith firm and then subtract from this aggregate the deduction that the same ith firm received that year. In the case of non-beneficiary firms, the deducted amount was 0. Thus, this variable showed for each firm the total deductions given to other firms in a specific industry. The same approach was followed to create the displace- ment for region-specific variables and for industry- and region-specific variables. The following formulas give the formal notation for the displacement variables as described:
51 Helge Bennmarker, Erik Mellander, and Björn Öckert, “Do Regional Payroll Tax Reductions Boost Employment?” (2009) 16 Labour Economics 480 - 89. 52 Bordignon et al., supra note 11. 53 Kangasharju, supra note 23. 20 n canadian tax journal / revue fiscale canadienne (2020) 68:1
N DIi,y = S i = 1( Di, y, k) - Di, y , i ∈ k, (7a)
N DRi,y = S i = 1( Di, y, r) - Di, y , i ∈ r, and (7b)
N DIRi,y = S i = 1( Di, y, k) - Di, y , i ∈k, r, (7c) where i represents firms,k indicates industry, r reflects region,y stands for year, and D is the amount of deductions. The estimation was implemented utilizing the DiD model described previously (equation 4) and including the displacement variables DI (for industry-level effect),DR (for regional-level effect), andDIR (for industry and regional effect). Finally, to detect firm relocation toward more-subsidized geographical zones, we analyzed the effect of the provision on net firm inflow at the “nomenclature of territorial units for statistics” (NUTS) 3 level (Italian provinces) by estimating the following equation over the provision’s duration period, 2005 - 2007:
Net_inflowp = a + b1Dedp + b2 A_Dedp + b3Zone × A_Dedp + e, (8) where Dedp is the total amount of deduction for the province p; A_Dedp is the aver- age deduction calculated for beneficiary firms; and Zone × A_Dedp is an interaction term between the more-subsidized zone and the average deduction.
MATCHED SAMPLE Differences in characteristics between beneficiary and non-beneficiary firms before deductions could result in biased estimates of the causal effects of access to the de- duction. This is because it is difficult to distinguish whether the performance of firms in those post-deduction years was attributable to the deduction itself or to the fact that firms with high performance tended to be beneficiaries. To determine the firm-specific characteristics that may affect the probability of being a tax-deduction-receiving firm, we referred to some prior studies of local job- creation incentive programs.54 The main variables used in these studies included tax status, age, age2, ROA, firm location (headquarters), size (in terms of persons em- ployed), a sales class dummy, and an economic activity dummy. We also added the ratio of tangible assets to total assets (Tangibles) and tendency to export (Exp_tend). Having or not having corporate tax (IRES) liabilities determined, in part, the benefits to firms from participating in the employment tax deduction. Firms that had low or no corporate tax liabilities had little incentive to take the IRAP deduction, while firms with high corporate tax liabilities in the current year were more likely to try to decrease liabilities by taking the deduction. Since the IRAP deduction facilitates the participation of firms in less-developed regions, those firms may be
54 Dagney Faulk, “The Participation of Firms in Tax Incentive Programs” (2001) 31:1 Review of Regional Studies 39 - 50; and Faulk, supra note 12. the evaluation of job tax incentives: an analysis of a regional tax n 21 more likely to take the deduction for job creation, and thus to participate in the tax incentive provision. The sales class is a measure of firm size. A tax incentive for larger firms may yield better outcomes than provision of the same incentive to smaller firms. Moreover, firms in certain industries may be more likely to take the tax deduction, either because certain industries use labour-intensive technologies and therefore can more easily increase employment, or because they are more likely to have corporate tax liabilities. To evaluate different specifications, we used the balancing condition in order to ensure that each independent variable did not differ significantly as between treated and non-treated firms. Thus, only treated and non-treated firms with the same propensity score and the same distribution of their observable characteristics were matched. Table 4 shows the results from estimating the probit model evalu- ated for the years before the deduction was introduced (that is, in the year t − 1). It appears that more capital-intensive firms were less likely to be beneficiaries, while more profitable firms and firms with corporate tax liabilities were more likely to be beneficiaries. A range of specifications with different satured models and interactions of differ- ent explicative variables were estimated.55 The outcomes were not significantly different from those of the base model, and there was no substantial improvement of the goodness-of-fit statistics. The base model was chosen because of its simplicity. The result of the matching procedure with regard to the common support is illustrated in figure 2. It can be observed that the common support extended over all propensity score values. To verify the quality of matching, t-tests for equality of means in the treated and non-treated groups, both before and after matching, were performed. For good balancing, t-statistics should be non-significant after matching. Moreover, the stan- dardized bias, as stated in Rosenbaum and Rubin,56 should be less than 5 percent after matching. As reported in table 5, all covariates were well balanced (with a percentage bias after matching of less than 5 percent). Thus, the matching proced- ures were effective in building a good control group. A subsample consisting of only matching units was also considered, reducing the sample size to around 347,000 firms. This set of firms was utilized to estimate the DiD model described in equation 5. The distribution of the control group firms by propensity score values, calcu- lated in the t − 1 year, is presented in figure 3.
55 The models’ estimations are available on request. 56 Paul R. Rosenbaum and Donald B. Rubin, “Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score” (1985) 39:1 American Statistician 33 - 38 (https://doi.org/10.2307/2683903). 22 n canadian tax journal / revue fiscale canadienne (2020) 68:1
TABLE 4 Probit Model To Estimate Propensity Score
Probability of being a beneficiary t( = 1)
Persons employed ...... 0.005 0.010 Age ...... −0.017 *** 0.000 Age 2 ...... 0.0001 *** 0.000 Tax status ...... 0.068 *** 0.008 ROA ...... 0.004 *** 0.000 Tangibles ...... 0.034 ** 0.014 Exp_tend ...... −0.080 *** 0.021 Productivity ...... −0.0004 *** 0.000 Size dummies ...... yes Area dummies ...... yes Industry dummies ...... yes LR chi2(22) ...... 8,292.4
Notes: Persons employed in thousands; productivity in thousands of euros. Standard error in italics. *** Statistically significant at the 0.01 level. ** Significant at the 0.5 level. * Significant at the 0.1 level.
TABLE 5 Balance Checking Statistics
Unmatched / Mean Reduction t-test Variable matched treated Control Bias (%) (%) bias t p > t
Persons employed . . . . U 31.912 22.728 3.4 7.64 0 M 31.912 29.565 0.9 74.4 1.37 0.171 Age ...... U 13.99 15.55 −13 −24.56 0 M 13.99 14.04 −0.4 96.9 −0.6 0.547 Age2 ...... U 332.97 393.52 −8.9 −16.39 0 M 332.97 333.81 −0.1 98.6 −0.18 0.854 Tax status . . . . U 0.83 0.8 8.5 16.07 0 M 0.83 0.83 −0.1 98.5 −0.19 0.846 ROA ...... U 7.55 6.66 9.2 17.29 0 M 7.55 7.53 0.2 97.6 0.31 0.756 Tangibles . . . . U 0.19 0.2 −5.3 −9.69 0 M 0.19 0.19 −0.1 98.0 −0.16 0.876 Exp_tend . . . . U 0.04 0.04 0.6 1.08 0.281 M 0.04 0.04 −0.6 −0.6 −0.8 0.423 Productivity . . . U 47.6 46.4 1.4 2.4 0.016 M 47.6 48.3 −0.9 35.2 −1.43 0.151 the evaluation of job tax incentives: an analysis of a regional tax n 23
FIGURE 2 After-Matching Estimated Probability of Being an IRAP Deduction Beneficiary (Propensity Score)—Distribution by Treated and Control Groups
10
5 Kdensity − pscore
0 0 0.1 0.2 0.3 0.4 Propensity score Treated Control
IRAP = regional tax on productive activities.
FIGURE 3 Distribution of the Control Group Firms by Propensity Score Values 3,500
3,000
2,500
2,000
No. of firms 1,500
1,000
500
0 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16 0.17 0.18 0.19 0.20 0.21 0.22 0.23 0.25
p score value 24 n canadian tax journal / revue fiscale canadienne (2020) 68:1
DIFFERENCE-IN-DIFFERENCE RESULTS Results To study whether the IRAP deduction has had any effects on employment in post- deduction years, we estimate the regression model in equation 5. The dependent variables are employment, sales, average salary, and productivity changes at the firm level, and the key estimate is the DiD estimator. Table 6 presents the effects of the IRAP deduction on post-deduction employ- ment estimated by means of an ordinary least square (OLS) robust estimator. The results for OLS estimation of the base model are shown in column 1. The
DiD estimator inter (TCAi × Aftert + s) is positive and indicates that, on average, the IRAP deductions had a positive effect on employment in the years for which the deduction was granted. The coefficients are also highly significant when firm-level controls—firm sales and the profitability ratio—are added (column 2). If we want to check the parallelism assumption (as described above) in the frame- work of the DiD identification strategy, we have to add an individual-specific time 57 trend to the regressors in Xst, as shown in Angrist and Pischke. If the estimated effects of interest remain substantially unchanged and significant by the inclusion of this trend, we can accept the results obtained by the DiD procedure described previ- ously. Column 3 in table 6 presents the estimated inter coefficients for theOLS base model when trend is included, and we can note that the inter coefficient (our DiD effects) registers a decrease but remains strongly significant; this confirms the effects found using our model. The DiD estimator for the model in column 3 suggests that the deduction has had a positive and significant impact on employment in beneficiary firms. These results are in line with the outcomes of, for instance, Huttunen et al.,58 although they present higher values and so a larger economic impact of the deduction on employment growth. To investigate the dynamic pattern of the post-deduction employment effects, in column 4 the interaction variable for the whole post-deduction period inter =
(TCAi × Aftert + s) with year-by-year interaction variables is replaced—that is, intera 0 = (TCAi × Aftert + 0 ), starting from the first deduction year and continuing for the next three years. All coefficients on these interactions are significant and present an increasing effect over time until the second year after the year of initial deduction. The third year presents a decrease relative to the second year. The differential outcome in terms of employment growth between treated and non-treated firms remains significant in all three periods after the last deduction year. Also, after 2008, when the financial crisis caused employment to begin to decrease, the outcomes of the non-beneficiaries do not fill the gap, but a decrease in the differential can be registered. This raises some doubt about the persistency of the deduction’s effects.
57 Angrist and Pischke, supra note 44. 58 Huttunen et al., supra note 6. the evaluation of job tax incentives: an analysis of a regional tax n 25 *** ** 0.406 0.742 (7) yes yes yes yes yes − 0.377 − 1.993 -ratio) OLS 296.70 508.6 0.121 t ( 346,807 − 173.742 Productivity − 6,420.11 *** * (6) yes yes yes yes yes 0.112 0.0002 0.000 0.188 2.146 0.099 -ratio) OLS 0.112 salary t ( − 0.0001 − 0.001 Average Average 346,807 *** *** (5) 0.681 1.080 2.540 yes yes yes yes yes -ratio) OLS Sales 0.147 t − 0.253 110.340 181.4 ( 346,807 − 30.373 − 607.25 *** *** *** *** *** *** (4) yes yes yes yes yes 0.145 0.001 0.001 0.001 0.001 0.001 1.989 0.007 0.113 0.026 0.047 0.062 0.051 -ratio) OLS 0.247 0.132 t ( 346,807 ** *** (3) yes yes yes yes yes 0.308 0.006 0.566 0.001 1.459 0.210 0.037 -ratio) OLS 0.112 t ( − 0.0002 346,807 gnificant at the 0.5 level. * Significant 0.1 Robust estimations. *** *** Employment no (2) yes yes yes yes 0.7 0.006 0.455 0.06 2.733 0.217 0.038 -ratio) OLS 0.097 t − ( − 0.0001 346,807 *** *** *** no no (1) yes yes yes 3.17 0.64 0.007 0.472 2.758 0.002 0.198 0.04 -ratio) OLS 0.111 t ( 346,807 2004-2010 and Productivity, Salary, Sales, Average Changes in Employment, Deduction on Post-Deduction . . . . . Effects of the IRAP 2 R . -statistics in italics. *** Statistically significant at the 0.01 level. ** Si T ...... Inter After_s Intera 0 TABLE 6 TABLE Treated Constant Firm controls Intera 1 Intera 2 Intera 3 Variables (changes) Variables No. of observations IRAP = regional tax on productive activities; OLS ordinary least square. Notes: Regional dummies Year dummies Year Industry dummies . Individual-specific trend . Adjusted 26 n canadian tax journal / revue fiscale canadienne (2020) 68:1
Our data permit an indirect evaluation of the quality of the jobs created. The evaluation is carried out by estimating the impact of the deduction on average salary and labour productivity measured at the firm level, as previously done by Monte- duro et al.59 Average salary could also be interpreted as a proxy of the firm’s human capital, and competitiveness increases if the ratio of highly qualified workers rises. A higher number of qualified workers also augments the efficiency of the firm, thus stimulating its growth. Arrighetti and Lasagni60 found a positive relationship be- tween the high level of human capital and the augmented probability of new hiring among Italian firms. We analyzed the impact of the deduction on average labour costs and labour productivity by treating those factors as dependent variables in equation 5. Columns 5 to 7 in table 6 show the results for these estimations. The analysis shows that the average effect of the deduction on the firms’ average salaries (column 6) was negative and non-significant—that is, there were no signifi- cant effects of treatment on average salaries. This result may be cautiously interpreted as an insufficient stimulus to hire more skilled employees. It may also suggest that the jobs created were of low quality (or at least the deduction did not help to enhance job quality). This result is consistent with the estimated effect of the deduction on productivity; in the latter case, the deduction had a strong negative impact (column 7). If there were no other elements (such as investments in terms of both tangible/intangible assets and human capital) to offset the increasing employ- ment, in general the provision of the deduction entailed a decrease in productivity. Despite this result, it seems that the deduction has had a positive and significant impact on sales (column 5), meaning that the provision may enhance firm growth. We similarly investigated the differential effects of treatment among the 11 Italian macro-regions.61 For each macro-region, we included the interaction effect between regional dummy and the treatment. The estimates are presented in table 7. The parameters are all statistically significant and positive except for the north- west macro region (comprising Piedmont, Liguria, and Valle d’Aosta). On average, the coefficients were higher for north-central regions, indicating a stronger impact for these regions. Despite the greater effort made toward rebalancing the employ- ment levels in southern regions (through larger deductions), it seems that the outcomes were limited.
59 Monteduro et al., supra note 21. 60 Alessandro Arrighetti and Andrea Lasagni, Assessing the Determinants of High-Growth Firms in Italy, Working Paper 2010 - 07 (Parma: University of Parma, Department of Economics, September 2012). 61 In this case, we utilized level 1 of the 1999 version of the Eurostat-NUTS classification (http:// ec.europa.eu/eurostat/web/nuts/history). the evaluation of job tax incentives: an analysis of a regional tax n 27
TABLE 7 Effects of the IRAP Deduction on Post-Deduction Employment at the Macro-Regional Level, 2004-2010
OLS OLS Variable (t-ratio) Variable (t-ratio)
After_s 0.237 1.1
North-Central South
North-west ...... −0.017 Abruzzi-Molise 0.027 *** −1.9 11.6 Lombardy ...... 0.036 *** Campania 0.022 *** 91.4 23.6 North-east ...... 0.042 *** South 0.021 *** 30.4 16.5 Emilia-Romagna ...... 0.041 *** Sicily 0.034 *** 92.9 27.6 Centre ...... 0.027 *** Sardinia 0.016 *** 43 Lazio ...... 0.052 *** 95.5 Firm controls ...... yes Year dummies ...... yes Individual-specific trend . . . yes Adjusted R2 ...... 0.141 No. of observations . . . . . 346,807
IRAP = regional tax on productive activities; OLS = ordinary least square. Notes: T-statistics in italics. *** Statistically significant at the 0.01 level. ** Significant at the 0.5 level. * Significant at the 0.1 level. Robust estimations.
Robustness Check We implemented a robustness check by calculating a DiD matching estimator in equation 6. In all estimates, we controlled for changes in the firm-specific charac- teristics of sales and ROA. The estimates were calculated for the three financial years after the last year for which the firm benefited from the deduction. The results, shown in table 8, indicate that the obtained DiD estimates were robust for almost all of the firms’ characteristics considered in our analysis. The ATT coefficients were not significant only for average salary. This suggests that there may have been weak (or null) effects of the deduction in terms of average salary (that is, in terms of the quality of the jobs created). Finally, the ATT estimates confirm a persistent effect of the deduction for some periods after the provision expired.
Displacement Effects The displacement effect for employment change was tested by adding each dis- placement variable in equation 7. If there was a displacement effect, the previously 28 n canadian tax journal / revue fiscale canadienne (2020) 68:1
TABLE 8 Effects of the IRAP Deduction on Employment, Productivity, Average Salary, and Sales—ATT Approach
Standard Variables ATT error t
Employmentt + 1 ...... 0.006 0.001 6.653 *** Employmentt + 2 ...... 0.007 0.001 9.093 *** Employmentt + 3 ...... 0.007 0.001 7.158 ***
Productivityt + 1 ...... −1.789 0.626 −2.861 *** Productivityt + 2 ...... −1.442 0.621 −2.322 *** Productivityt + 3 ...... −1.463 0.625 −2.34 ***
Average salaryt − 1 ...... 0.107 0.184 0.584 Average salaryt − 2 ...... 0.082 0.193 0.424 Average salaryt − 3 ...... 0.231 0.207 1.112
Salest + 1 ...... 804 296 2.716 *** Salest + 2 ...... 874 327 2.671 *** Salest + 3 ...... 874 352 2.483 *** ATT = average effect of treatment on treated; IRAP = regional tax on productive activities. Notes: *** Statistically significant at the 0.01 level. ** Significant at the 0.5 level. * Significant at the 0.1 level. Standard errors are obtained by bootstrapping (100 iterations). found coefficient should be influenced by the impact on other firms, as studied by Kangasharju.62 The analysis was carried out considering the deduction amounts for region-level and industry-level data. Results give no indication of a displacement effect that may bias the estimate of deduction effects, even if the interaction effect between industries and regions was considered. The estimated results of deductions were in all cases positive, but none indicated a statistically significant effect. The results for firms’ net inflow estimated by equation 8 did not present signifi- cant effects of theIRAP deduction’s differential treatment between the north-central and southern zones. This indicates that the deduction amount had no effect on the creation of new firms and did not provide an incentive to relocate firms to zones with more generous deduction schemes. The results of our analysis of displacement effects encourage us to consider that there was little interference owing to such effects and the stable unit treatment value assumption is satisfied.
CONCLUSIONS This article has provided estimates of the employment impact of the Italian IRAP deduction granted for new additional jobs created in the period 2005 - 2007, by comparing employment (and other variables) of firms that exploited the deduction and firms that did not exploit the deduction. Other studies analyzing the effects of
62 Kangasharju, supra note 23. the evaluation of job tax incentives: an analysis of a regional tax n 29 economic policy measures have utilized, as an outcome variable, the target of the policy provisions itself; however, we have further broadened the study’s objectives to include the verification of other aspects, such as duration of the employment increase and effects on employment structure of the firm. Results from the models indicate that firms taking advantage of the IRAP tax incentive created relatively more employment opportunities; but, more relevant, they also maintained this new employment at least up to 2008. Moreover, the effi- ciency analysis revealed that these provisions performed well in terms of cost per job created. These results are noteworthy also because the analysis was conducted on a large sample of firms. Moreover, the analysis yielded results that are quantitatively similar to those of other studies relevant to this topic, as described above. Other important aspects that we tried to verify are related to the effects of the provision both on other important economic variables and at the regional level. The effects of the provision in terms of job quality seem to be negative, since average salaries did not show significant changes and productivity presented a negative effect for beneficiaries. Perhaps firms did not accurately calculate the optimal amount of new employment required to obtain the deductions, and thus restricted growth in efficiency and productivity, as suggested by Bernini and Pellegrini.63 Only the growth in volume of sales presented a positive and significant differential. Efforts to redirect new employment toward the southern regions do not seem to have been successful. The evidence derived from this kind of analysis may help in building forecasting models to foresee the effect of a fiscal provision and its interaction with the eco- nomic cycle. These models can illuminate the economic effects of tax policies on firms and the influence of such policies on firms’ competitiveness. This information should be provided to policy makers so that they may know the expected impact before designing incentive programs.
APPENDIX EFFICIENCY ESTIMATION Having produced estimates regarding the effectiveness of the IRAP deduction, we can determine the efficiency of the provision. For this purpose, we transformed the impact estimates obtained from the econometric analysis into more easily inter- preted quantities. Following the approach utilized for other Italian subsides (as discussed, for example, in the ASVAPP report),64 we added cost information to the firm variables, thus obtaining a simple cost-effectiveness measure—the average cost per job created—calculated as follows:
63 Cristina Bernini and Guido Pellegrini, “How Are Growth and Productivity in Private Firms Affected by Public Subsidy? Evidence from a Regional Policy” (2011) 41:3Regional Science and Urban Economics 253 - 65. 64 ASVAPP, supra note 47. 30 n canadian tax journal / revue fiscale canadienne (2020) 68:1
To t al funds disbursed to beneficiary firms Cost per job created = ______ . Average employment impact × no. of beneficiaries
Analogously to employment, we present the impact on sales in the following form:
To t al funds disbursed to beneficiary firms Cost per extra euro of sales = ______ . Average impact on sales × no. of beneficiaries Some firms may exhibit a small number of jobs created, but they may also receive relatively fewer resources. For example, larger firms receiving larger grants tend to generate larger gains in employment per firm, but they also receive huge amounts of money, so the cost of job creation is also much higher. As shown in table A.1, the cost-effectiveness indicators (cost of a job) are not high. This may be interpreted as a sign of efficiency if these results are compared with those reported in other studies of similar provisions.65 The average costs are higher in the services sector and for larger enterprises. This is a reasonable result given the higher amount of deductions granted to larger enterprises. The cost per extra euro of sales is quite high. Thus, in this case, it is difficult to take this result as a sign of efficiency; the cost is also high relative to the effects of other provisions. Moreover, the cost dramatically increases for services. One possible objection is that a large fraction of the deduction amount is in- creasingly concentrated in southern Italy, which enjoys a higher deduction for each job created. The efficiency performance of firms associated with the IRAP deduction is marked by highs and lows. The cost per job created is not high with respect to other similar provisions that aim to encourage employment, at both national and regional levels.66 The cost per extra euro of sales otherwise indicates a poor efficiency performance. Finally, we must highlight that this analysis most likely overestimated the costs. We could not consider the positive feedback effects, including higher consumption and increased fiscal revenues driven by salary increases resulting from additional employment.
65 Ibid. 66 See, for example, ibid. the evaluation of job tax incentives: an analysis of a regional tax n 31
TABLE A.1 Average Impacts and Cost-Effectiveness of the IRAP Deduction on Employment and Sales, by Economic Activity and Firm Size
No. of supported firms Variable used in the analysis Costa
Employment ...... Total ...... 26,088 3,891 Industry ...... 11,637 3,730 Services ...... 14,451 6,304 Turnover class Less than € 2 million ...... 14,118 3,103 € 2 million to € 10 million ...... 8,954 10,842 Over € 10 million ...... 3,016 18,264 Sales ...... Total ...... 26,088 41 Industry ...... 11,637 14 Services ...... 14,451 166
IRAP = regional tax on productive activities. Note: Figures not reported in the case of negative coefficient estimates or results with no statistical significance at the level of 0.1. a Employment: cost per job created. Sales: cost per extra euro of sales. canadian tax journal / revue fiscale canadienne (2020) 68:1, 33 https://doi.org/10.32721/ctj.2020.68.1.pf.editor
Policy Forum: Editor’s Introduction—Taxes and Spending in Canada’s 43rd Parliament
The October 2019 Canadian federal election resulted in the return of the Liberal government, but now without a majority of seats in Parliament. Given this elec- toral outcome, what are the implications for tax and spending decisions? We have brought together two articles to address these fiscal questions in this Policy Forum. While the authors are now in academia, they all have experience advising govern- ments in the past. That experience provides necessary insight into not just what ought to be done but also what can be achieved, given administrative and political constraints. The first article is from Sean Speer. Speer provides specific examples oftax policy changes that may find favour in this Parliament. He also delivers a more fundamental structural message about the tax reform process. Building on the previous Policy Forum on the topic,1 Speer argues that successful tax reform must engage normative questions that require political input and deliberation. This is in contrast to a “technocratic” model that attempts to isolate reform from democratic input and pressures. Speer believes that there are good prospects for cooperation on a middle-income tax cut and a substantial program and tax expenditure review. In the second article, Rob Gillezeau and Trevor Tombe discuss the prospects for expenditure decisions by the 43rd Parliament. They provide detailed and insightful analysis of the expenditure plans in the Liberal platform, as well as in the plans of the opposition parties, whose support will be needed to pass budgets and legisla- tion. The analysis, however, goes beyond immediate policy initiatives by placing the discussion in the context of larger themes: long-run fiscal sustainability, and the economic appropriateness and interplay of federal-provincial spending. Together, these two articles show that there is a clear and implementable pro- gram on both taxes and expenditures that should be feasible, even with a minority government. Kevin Milligan Editor
1 See “Policy Forum” (2018) 66:2 Canadian Tax Journal 349 - 99.
33 canadian tax journal / revue fiscale canadienne (2020) 68:1, 3 5 - 4 7 https://doi.org/10.32721/ctj.2020.68.1.pf.speer
Policy Forum: Tax Reform in Canada’s 43rd Parliament—Politics, Policy, and Second-Best Choices
Sean Speer*
PRÉCIS Un Parlement minoritaire à Ottawa demandera du gouvernement et des partis d’opposition qu’ils trouvent des domaines de convergence politique. Un de ces domaines pourrait être la prise de mesures pour effectuer un examen complet du système fiscal fédéral, ce à quoi s’étaient engagés plusieurs partis politiques dans leur campagne électorale. Il est possible de procéder à un exercice d’examen qui améliorera en fin de compte l’efficacité et l’équité du code des impôts fédéral, mais un tel examen devra reconnaître qu’on ne peut pas séparer la politique fiscale de la politique. Cet article présente une analyse critique des appels à la création d’une commission royale ou à l’adoption du modèle big bang de réforme fiscale. Ces approches à la réforme fiscale supposent que la politique est un obstacle majeur à une réforme fiscale fondée sur des données probantes et que les impulsions politiques doivent être minimisées ou exclues du processus. Cette perspective ne tient pas compte des aspects normatifs de la politique fiscale et de la mesure dans laquelle les choix politiques sont façonnés par un mélange complexe d’intérêts, de préférences et de valeurs. L’auteur plaide plutôt pour un exercice de réforme qui s’enracine dans les institutions et les processus politiques. Il propose en particulier un modèle progressif mais systématique de réforme fiscale qui s’inspire de l’expérience des « examens stratégiques » des dépenses de programmes au fédéral, menée de 2007 à 2011. Dans ce modèle, le gouvernement procéderait à des examens thématiques des dépenses fiscales (axés, par exemple, sur l’accession à la propriété, la retraite et l’épargne, et les investissements dans les énergies propres) sur une base régulière et continue, afin de rationaliser progressivement le système fiscal fédéral en consolidant et en redéfinissant les dépenses fiscales actuelles. Les résultats peuvent être moins transformateurs à court terme que ceux qui pourraient être obtenus par d’autres approches, mais ils conduiront en fin de compte à une réforme plus durable à long terme.
* Of the Munk School of Global Affairs and Public Policy, University of Toronto (e-mail: [email protected]); previously an adviser to the Right Honourable Stephen Harper.
35 36 n canadian tax journal / revue fiscale canadienne (2020) 68:1
ABSTRACT A minority Parliament in Ottawa will require that the government and the opposition parties search for areas of policy convergence. One area where there is potential for cooperation is action on a comprehensive review of the federal tax system, which was an election commitment by multiple political parties. There is scope for a review exercise that ultimately can enhance the efficiency and equity of the federal tax code. But such a review will need to recognize that tax policy and politics cannot be divorced from one another. This article presents a critical analysis of calls for a royal commission or adoption of the “big-bang” model for tax reform. These approaches to tax reform assume that politics is a major barrier to evidence-based tax reform and that political impulses need to be minimized or excluded from the process. This perspective neglects the normative aspects of tax policy and the extent to which policy choices are shaped by a complex mix of interests, preferences, and values. Instead, the author argues for a reform exercise that is rooted in political institutions and processes. In particular, he proposes an incremental yet systematic model for tax reform that draws from the 2007- 2011 experience with “strategic reviews” of program spending at the federal level. This model would envision the government conducting thematic reviews of tax expenditures (focusing, for example, on home ownership, retirement and savings, and clean energy investments) on a regular, ongoing basis, in order to gradually rationalize the federal tax system by consolidating and redesigning current tax expenditures. The results may be less transformative in the short term than those potentially achievable by alternative approaches, but will ultimately lead to more durable reform over the long term. KEYWORDS: TAX REFORM n TAX POLICY n TAX EXPENDITURES n EFFICIENCY n EQUITY n POLITICAL ECONOMY
CONTENTS Introduction 36 The Case for Tax Reform 38 Is It Time for a Royal Commission? 39 Conditions for Successful Tax Policy Reform 41 The Strategic Review Model—An Alternative to a Royal Commission 44 Conclusion 47
INTRODUCTION As we approach the 2020 federal budget, there is plenty of speculation about how the Trudeau government will navigate the interplay between politics and policy that is inherent to budget making in a minority parliament. If “politics is the art of the possible,” as Otto von Bismarck famously observed, designing a budget for a min- ority context is often an exercise in second-best theory.1 It will invariably involve a series of political trade-offs.
1 R.G. Lipsey and Kelvin Lancaster, “The General Theory of Second Best” (1956) 24:1 Review of Economic Studies 11 - 32 (https://doi.org/10.2307/2296233). policy forum: tax reform in canada’s 43rd parliament n 37
In October 2019, Canadians elected a Parliament with a strong Liberal plurality but without sufficient votes to pass legislation or maintain the confidence of the House without support from other parties. Identifying policy and political con vergence will therefore be key to the government’s survival. The good news is that there are some areas of broad political convergence that can be the subject of cross- party support in the 43rd Parliament. Perhaps the most significant example is tax policy. The major political parties differed on various aspects of tax policy in the 2019 campaign, but there were also areas of considerable overlap. There was some com- bination of support for several policy changes, including lowering personal income taxes for low-income earners,2 increasing the generosity of the volunteer firefighter tax credit,3 cracking down on tax loopholes,4 and modernizing the tax treatment of non-domiciled digital service providers.5 The most interesting area of political convergence on tax policy, however, may be the Liberal and Conservative parties’ agreement on the need for a review of the federal tax code. There is certainly scope for such an exercise, as previous Canadian Tax Journal articles have observed.6 The growing number of tax expenditures has complicated the tax system and in turn made it less efficient and less progressive. A tax reform exercise that sought to achieve rationalization could therefore conceiv- ably achieve both conservative and progressive ends. But, as this article argues, tax reform will not occur according to the “big-bang” model favoured by some politicians, business organizations, and policy experts. This approach, including the creation of a royal commission, fails to reckon with the extent to which tax policy is informed and shaped by normative factors that can- not be reconciled through technocratic processes or sweeping changes. Instead, tax reform will need to come in the form of an incremental yet systematic model. One might think of it as an ongoing exercise of second-best choices rooted in theory and practice.
2 Liberal Party of Canada, Forward: A Real Plan for the Middle Class (Ottawa: Liberal Party of Canada, 2019), at 7 (https://2019.liberal.ca/wp-content/uploads/sites/292/2019/09/Forward -A-real-plan-for-the-middle-class.pdf ); and Conservative Party of Canada, Andrew Scheer’s Plan for You To Get Ahead (Ottawa: Conservative Party of Canada, 2019), at 3 (https://cpc-platform .s3.ca-central- 1.amazonaws.com/CPC_Platform_8.5x11_FINAL_EN_OCT11_web.pdf ). 3 Andrew Scheer’s Plan for You To Get Ahead, supra note 2, at 8; and New Democratic Party of Canada, A New Deal for People: New Democrats’ Commitments to You (Ottawa: NDP, 2019), at 87 (https://action.ndp.ca/page/-/2019/Q2/2019 - 06 - 19_Commitments-Doc_EN.pdf ). 4 Forward: A Real Plan for the Middle Class, supra note 2, at 79 - 80; and A New Deal for People: New Democrats’ Commitments to You, supra note 3, at 44. 5 Forward: A Real Plan for the Middle Class, supra note 2, at 79; Andrew Scheer’s Plan for You To Get Ahead, supra note 2, at 100; and A New Deal for People: New Democrats’ Commitments to You, supra note 3. 6 See, for example, Fred O’Riordan, “Policy Forum: Why Canada Needs a Comprehensive Tax Review” (2018) 66:2 Canadian Tax Journal 351 - 62. 38 n canadian tax journal / revue fiscale canadienne (2020) 68:1
The first section of the discussion that follows will examine the case for tax reform. The second section will consider the limits of an arm’s-length tax review process, including the possible creation of a royal commission, owing in large part to the inherent politics of tax policy. The third section will describe the political economy conditions for successful tax policy reforms. The fourth and final section will set out the case for an incremental yet systematic approach to tax reform that can make steady and ongoing progress in the direction of rationalization.
THE CASE FOR TAX REFORM It is not surprising that the Liberal Party and the Conservative Party have both committed to a review of the federal tax system. The idea has been percolating in Ottawa for several years. It is the subject of growing support among stakeholders that include the Business Council of Canada7 and the Canadian Chamber of Com- merce,8 Senate committees,9 and other high-profile voices.10 Some have even argued for the establishment of a royal commission.11 What is the motivation for tax reform? The Liberal Party and the Conservative Party have different reasons behind their platform commitments. The Liberal Party’s proposal was framed primarily around enhancing progressivity. The Conservative Party’s case for such a review was mostly focused on improving economic competitiveness. Yet these rationales are not necessarily incompatible. Most stakeholders who have advocated for such a review seem to be motivated by the goals of tax simplification and broad-based tax reform (involving a broader base and flatter rates) that could conceivably improve both efficiency and equity.
7 Jesse Snyder, “Business Group Calls for Sweeping Tax Reforms Ahead of Federal Election as Competitiveness Worries Deepen,” National Post, April 25, 2019 (https://nationalpost.com/ news/politics/business-group-calls-for-sweeping-tax-reforms-ahead-of-federal-election-as -competitiveness-worries-deepen). 8 Canadian Chamber of Commerce, “Time To Cut Canada’s Losses on Obsolete Tax System, Says Canadian Chamber of Commerce,” News Release, October 4, 2019 (www.chamber.ca/ media/news-releases/Time_to_cut_Canada%27s_losses_on_obsolete_tax_system). 9 See, for example, Canada, Senate, Fair, Simple and Competitive Taxation: The Way Forward for Canada: Report of the Standing Senate Committee on National Finance (Ottawa: Senate Standing Committee on National Finance, December 2017) (https://sencanada.ca/content/sen/ committee/421/NFFN/Reports/NFFN_Tax_Planning_24th_Report_e.pdf). 10 For example, Alan Lanthier, “The $40 -Billion Reason Canada Needs Real Corporate Tax Reform,” Financial Post, July 16, 2019 (https://business.financialpost.com/opinion/the- 40 -billion-reason-canada-needs-real-corporate-tax-reform). 11 Gerry Macartney, “Canada Needs a New Royal Commission on Taxation,” London Free Press, February 22, 2019 (https://lfpress.com/opinion/columnists/macartney-canada-needs-a-new -royal-commission-on-taxation); Mike Holden, Restoring Canada’s Advantage: A Need for Tax Reform (Canadian Manufacturers & Exporters and BDO, June 2018) (https://cme-mec.ca/ wp-content/uploads/2018/11/Doc_QC_Restoring-Canadas-Advantage.pdf ); and Jim Warren, “It’s Time for a Royal Commission on Taxation,” Toronto Sun, November 11, 2017 (https:// torontosun.com/opinion/columnists/warren-its-time-for-a-royal-commission-on-taxation). policy forum: tax reform in canada’s 43rd parliament n 39
The case for such reforms is certainly justified. It has been nearly 30 years since the federal government undertook a serious reform of the credits, deductions, and other special provisions included in the tax code. The result is a buildup of tax expenditures that complicate the tax system, narrow the tax base, and harm both efficiency and progressivity. The cost to government is also significant. Tax expenditures in the federal tax system alone now represent as much as $117.9 billion in forgone revenues or more than half of total revenues from income taxes and the goods and services tax (GST).12 A 2014 study by former Statistics Canada chief statistician Munir Sheikh estimated that the size of government in Canada when accounting for tax expendi- tures increased from 44 percent to 54 percent of gross domestic product.13 Yet while we have seen regular reviews of program spending over the past 25 years, the federal tax system has not been subjected to the same level of rigour. This amounts to effectively excluding from regular and ongoing scrutiny a set of federal policies that in total represent as much as three-quarters of direct program spending.14 There is a therefore good case for the federal government to launch a review of the tax system with the goal of making it more efficient, fairer, and simpler.
IS IT TIME FOR A ROYAL COMMISSION? The question of course is, how should such a review be conducted? The Policy Forum in a 2018 issue of this journal grappled with this question.15 The “how,” as Jennifer Robson put it in one of the issue’s articles, is as important as the “what” because the format and process will invariably shape the eventual policy outcomes.16 The prevailing view among contributors to that Policy Forum was that, on balance, comprehensive reform was preferable to incremental changes and an
12 Canada, Department of Finance, Report on Federal Tax Expenditures—Concepts, Estimates and Evaluation 2017 (Ottawa: Department of Finance, 2017) (www.canada.ca/en/department- finance/services/publications/federal-tax-expenditures/2017.html). There are limits to summing up the cumulative costs of tax expenditures because of the interaction of such measures. Changes to the basic personal amount, for instance, would affect revenue costs of other tax credits. Still, the Department of Finance’s estimate of $117.9 billion is a good back-of-the- envelope indication of the magnitude of the federal government’s mix of tax expenditures. 13 Munir A. Sheikh, Estimating the True Size of Government: Adjusting for Tax Expenditures (Ottawa: Macdonald-Laurier Institute, February 2014) (www.macdonaldlaurier.ca/files/pdf/ MLISheikhPaper02 - 14 -final.pdf ). 14 Direct program spending (which excludes transfers to persons and other levels of government) is estimated to be roughly $152 billion in 2019 - 20: Canada, Department of Finance, 2019 Budget, Budget Plan, March 19, 2019, at 289, table A2.6. 15 “Policy Forum” (2018) 66:2 Canadian Tax Journal 349 - 99. 16 Jennifer Robson, “Policy Forum: Building a Tax Review Body That Is Fit for Purpose— Reconciling the Tradeoffs Between Independence and Impact” (2018) 66:2Canadian Tax Journal 375 - 86, at 377. 40 n canadian tax journal / revue fiscale canadienne (2020) 68:1 arm’s-length process was preferable to conventional policy making.17 Joseph Heath’s article, which proposed a permanent administrative agency that would have del- egated responsibility for setting tax policy similar to the Bank of Canada’s role in monetary policy, was the most decidedly in favour of such a technocratic model.18 As he wrote, “an ITA [independent tax authority] represents a response to a genuine problem, which is the exploitation by politicians of public ignorance and irrational- ity with respect to taxes.”19 The underlying assumption for these different models (including a royal com- mission) is that there is a need to “depoliticize” the review process in particular and tax policy more generally.20 The models vary by degree on how to achieve this goal. The common view, though, is that a technocratic process would insulate federal tax policy from the vicissitudes of politics. This perspective fails, in my view, to properly account for the normative dimen- sion of tax policy. The unit of taxation, the level of progressivity, the trade-off between efficiency and equity, and even definitions of income are informed and shaped by normative forces. Delegating these choices to economists or tax policy experts is asking them to act as moral philosophers.21 Of course, we can attempt to bring empirical analysis to bear in trying to address these questions. Analyses of the costs of taxation or the marginal efficiency costs of different forms of taxation,22 or of behavioural responses to changes in marginal tax rates,23 are useful inputs into the policy process. But it is misguided to think that data and evidence are all we need to craft an optimal tax policy framework. Most people, including economists, ultimately rely on a complex set of preferences to reconcile the inherent tensions in public policy. It is no surprise, therefore, that research finds that few of us actually hold strict utilitarian views on tax policy.24
17 The exception was Shirley Tillotson, whose contribution was a historical perspective on past tax reform efforts. She argued in favour of “incremental change” because of the inevitable role of politics. See Tillotson, “Policy Forum: Then and Now—A Historical Perspective on the Politics of Comprehensive Tax Reform” (2018) 66:2 Canadian Tax Journal 363 - 74, at 373 - 74. 18 Joseph Heath, “Policy Forum: From Independent Tax Commission to Independent Tax Authority” (2018) 66:2 Canadian Tax Journal 387 - 99. 19 Ibid., at 389. 20 Ibid., at 387. 21 Benjamin B. Lockwood and Matthew Weinzierl, “Positive and Normative Judgments Implicit in U.S. Tax Policy, and the Costs of Unequal Growth and Recessions” ( January 2016) 77 Journal of Monetary Economics 30 - 47. 22 Jason Clemens, Niels Veldhuis, and Milagros Palacios, “Tax Efficiency: Not All Taxes Are Created Equal” [2007] no. 4 Studies in Economic Prosperity (www.fraserinstitute.org/studies/ tax-efficiency-not-all-taxes-are-created-equal). 23 Canada, Department of Finance, Tax Expenditures and Evaluations 2010 (Ottawa: Department of Finance, 2010), at 45 - 65. 24 Matthew Weinzierl, “The Promise of Positive Optimal Taxation: Normative Diversity and a Role for Equal Sacrifice” (2014) 118Journal of Public Economics 128 - 42 (http://dx.doi.org/ 10.1016/j.jpubeco.2014.06.012). policy forum: tax reform in canada’s 43rd parliament n 41
If tax policy is messy and complicated, this is in large part because our competing normative perspectives on the underlying questions are messy and complicated. Decisions on tax policy are, in my view, less about exploitation by politicians of public ignorance25 and more about the work of politics to try to reconcile our dif- ferences. The outcome may be suboptimal, but optimality is a mostly conceptual question in a world of competing interests, preferences, and values.26 Reconciling these tensions is an inherently political project. It is even more so given that the government’s taxing power is a fundamental part of the social contract between individuals and the state.27 We grant the state a taxation capacity to ensure that the distribution of public benefits and public burdens is decided by collective political deliberation rather than by individual actions and choices. Gov- ernment is then responsible for the protection of property rights, production of public goods, and redistributive goals on our behalf. It is therefore wrong to think of taxation as merely a technical matter similar to monetary policy. It is more accurate to see it as foundational to our relationship with the government and how the state interacts with individuals and the broader society. Outsourcing responsibility for tax policy’s design, development, and imple- mentation, as Heath and others propose, neglects this fundamental point. The key takeaway then, in my view, is that efforts to divorce tax policy from politics are both conceptually mistaken and ultimately bound to fail. Politics is the rightful mechanism for tax policy questions.
CONDITIONS FOR SUCCESSFUL TAX POLICY REFORM This, of course, does not mean that the Liberal Party and the Conservative Party should abandon their commitments to carry out a review of the federal tax system. It just means that the process needs to lean into politics rather than try to exclude it. One way to judge how to best account for the role of politics is to think about conditions for successful tax policy reform. While this may seem counterintuitive, the best approach is to consider the characteristics of failed policy changes. There are, in my experience, three factors associated with unsuccessful tax reforms that can inform how to design a more effective process. It is worth outlining them here before developing a better model for the Trudeau government to deliver on its platform pledge.
25 As argued by Heath, supra note 18. 26 Sean Speer, “We Need To Restore Our Diminished Politics,” Policy Options Politique, September 18, 2019 (https://policyoptions.irpp.org/magazines/september- 2019/we-need-to -restore-our-diminished-politics). 27 Allison Christians, Sovereignty, Taxation, and Social Contract, University of Wisconsin Law School Legal Studies Research Paper no. 1063 (Madison, WI: University of Wisconsin Law School, August 2008) (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1259975). 42 n canadian tax journal / revue fiscale canadienne (2020) 68:1
The first factor is that complexity is a vice. Not only is it impossible for taxpayers to understand a complex set of changes; it is also challenging for policy makers to communicate them, and in turn becomes much easier for critics to set the prevailing narrative. Complexity also often leads to longer implementation timelines, which produce more political exposure.28 The second factor is that a government cannot have too many political fronts open at once. Otherwise the government risks creating the perception that it is embattled and increases the probability for a policy reversal. Reform gradualism, by contrast, can enable policy makers to be intentional about targeting different policy areas and taxpayer groups at different times.29 The third factor is that reform cannot be distributionally skewed. It is notable, for instance, that while the totality of the Harper government’s tax policies enhanced the tax system’s progressivity, the government was still criticized for its policy changes that benefited higher-earning households.30 That experience is a reminder that the current political environment will not permit a set of tax changes that are not carefully calibrated to achieve distributional balance or even enhanced progres- sivity. Such calibration can be more difficult when a government is attempting to manage several reforms involving various tax expenditures and even different forms of taxation.31 The recent US tax reform package in 2017 may present a counter example to this premise, but that episode might best be explained as a product of the United States’ unique political institutions and its dire need to reform the tax treat- ment of corporations. Proponents of “big bang” tax reform tend to argue that it is a better model than incremental change because it can more effectively deal with interactions among parts of the tax system—think, for instance, of the integration between personal and corporate income taxes—and provides policy makers with more tools to trade off efficiency and distributional considerations. These arguments are not without basis. In an idealized world, the “big bang” model would be the most efficient means of producing a better tax system. But democracy and pluralism are not necessarily prone to efficiency, especially in a world of competing preferences and competing views about the public interest. Those arguing in favour of the “big bang” model or of outsourcing reform deci- sions to non-political actors do not tend to satisfactorily engage the inherent role of politics in reconciling these differences in tax policy. This is a huge blind spot.
28 Bert Brys, Making Fundamental Tax Reform Happen, OECD Tax Working Papers no. 3 (Paris: OECD, 2011) (https://doi.org/10.1787/22235558), at 20. 29 Paola Profeta and Simona Scabrosetti, “The Political Economy of Taxation in Europe” (2017) 220:1 Hacienda Pública Española/Review of Public Economics 139 - 72 (www.ief.es/docs/destacados/ publicaciones/revistas/hpe/220_Art5.pdf ). 30 Trevor Shaw, Revenue and Distributional Analysis of Federal Tax Changes: 2005 - 2013 (Ottawa: Office of the Parliamentary Budget Officer, May 27, 2014) (www.pbo-dpb.gc.ca/web/default/ files/files/files/Fiscal_Impact_and_Incidence_EN.pdf ). 31 See Brys, supra note 28. policy forum: tax reform in canada’s 43rd parliament n 43
Recent debates about income splitting for two-earner families, or higher rates for top income earners, or carbon taxation, are not evidence of politics run amok. They are examples of politics working through nuanced questions involving a combina- tion of normative and empirical considerations. Experts, organizations, and scholars in favour of a royal commission (or some variation thereof ) also do not satisfactorily engage the political economy challenges associated with instituting sweeping, comprehensive reforms or outsourcing the process to non-political actors. This is another blind spot and is especially problem- atic in a populist moment when western societies are expressing skepticism about the wisdom of experts.32 The cultural, political, social, and technological environ- ment has fundamentally changed since the Carter commission.33 It stands to reason that we similarly need to adjust how we think about the right model for reviewing and reforming the federal tax system. Managing the risks of complexity, political exposure, and distributional effects is best achieved by an incremental yet systematic model developed through conven- tional policy making. A focused, targeted approach to tax reform enables policy makers to take on a discrete piece of the tax system and to refine and adjust its design in a relatively short time frame involving a clear and identifiable group of taxfilers. This model allows policy makers to navigate the normative waters of tax policy with fewer challenges than they face when aiming to take on the whole system at once. It will be less ambitious than some stakeholders would prefer and may require second-best choices. But if well-structured and properly executed, it can achieve steady and ongoing progress in the direction of rationalization. The goal should be to replicate the Trudeau government’s consolidation of care- giving tax expenditures in the 2017 budget34 across other parts of the tax system. The new Canada caregiver credit consolidated three separate tax expenditures related to caregiving, streamlined the eligibility rules, enhanced the generosity, and ultimately simplified the tax system. These changes were broadly supported by stakeholders as well as tax and social policy experts.35 The 2017 measures are a good example of politics and policy reinforcing one another. The government was able to advance a political priority related to care giving at minimal incremental cost because the three existing tax expenditures were
32 Tom Nichols, “How America Lost Faith in Expertise and Why That’s a Giant Problem,” Foreign Affairs, March/April 2017 (www.foreignaffairs.com/articles/united-states/2017 - 02 - 13/ how-america-lost-faith-expertise). 33 The Royal Commission on Taxation, which issued its six-volume report in 1966 - 67. 34 Canada, Department of Finance, 2017 Budget, Budget Plan, March 22, 2017, at 204 - 5. 35 See, for example, Sherri Torjman, Michael Mendelson, and Ken Battle, The 2017 Farewell Budget (Ottawa: Caledon Institute of Social Policy, March 2017) (https://maytree.com/ wp-content/uploads/1111ENG.pdf ); and Canadian Cancer Society, “Federal Budget Investment in Home and Palliative Care a Win for Cancer Patients,” Press Release, March 22, 2017 (www.cancer.ca/en/about-us/for-media/media-releases/national/2017/federal-budget -announcement/?region=on). 44 n canadian tax journal / revue fiscale canadienne (2020) 68:1 already costing roughly $185 million per year.36 And in turn we got a simpler, more progressive, and better-functioning tax system. The government should carry out similar thematic reforms across the federal tax system. It can tackle a specific set of policy themes each year and incrementally yet systematically make progress in the direction of rationalization through a combin- ation of consolidation and redesign.
THE STRATEGIC REVIEW MODEL—AN ALTERNATIVE TO A ROYAL COMMISSION What model could the government draw from? The Harper government’s experiment with regularized, annual reviews of pro- gram spending may provide a useful model. The “strategic review” process ran from 2007 to 2011 as a bottom-up exercise to scrutinize program spending across fed- eral departments.37 The goal was not primarily focused on fiscal savings but rather on controlling the growth of new spending. Strategic reviews were supposed to identify low-priority spending that could be reallocated to new, higher priorities. Roughly 25 percent of federal program spending was reviewed annually over a four-year cycle. Programs and services were subjected to various tests—including core federal role, efficiency, and priorities—and 5 percent of departmental spending was to be reallocated from low-performing, low-priority activities to higher ones. Ministers identified fiscal savings in their respective portfolios and were able to put forward proposals suggesting where these savings could be “reinvested” in their departments. Strategic reviews contributed to some useful reforms and helped to control the growth of new program spending. The results were reflected in the federal budgets from 2008 to 2011.38 In total, $2.8 billion in annual savings was realized and re- cycled to new and different priorities.39 There is scope to extend the strategic review model to the tax system. It would not be precisely the same for various reasons, including (but not limited to) the concentration of tax expenditures in the Department of Finance. But the govern- ment could fulfill its platform commitment by establishing a regularized review process that evaluates and reforms different components of the tax system on an annual basis.
36 The caregiver credit was $100 million, the family caregiver tax credit was $75 million, and the infirm dependant tax credit was $10 million. Canada, Department of Finance, Report on Federal Tax Expenditures—Concepts, Estimates and Evaluations 2016 (Ottawa: Department of Finance, 2016), at 34. 37 Treasury Board of Canada Secretariat, “Strategic Reviews” (www.tbs-sct.gc.ca/sr-es/index-eng .asp). 38 Kevin McCarthy and Sean Speer, “Supporting Ontario’s Fiscal Strategy,” Ontario 360, November 26, 2019 (https://on360.ca/policy-papers/supporting-ontarios-fiscal-strategy). 39 See “Strategic Reviews,” supra note 37. policy forum: tax reform in canada’s 43rd parliament n 45
Rather than targeting a share of spending, a strategic review process for the tax system could operate on a thematic basis similar to the consolidation of caregiving- related tax expenditures. This thematic model could be expanded to include home ownership, post-secondary education, employment, medical expenses, savings and retirement, and aging, as well as fossil fuels, clean energy investments, research and development, capital expenses, and small businesses. Consider home ownership, for instance. Currently the federal tax code contains at least six housing-related tax expenditures totalling over $7 billion per year in forgone revenues.40 Accepting that the government intends to continue support- ing housing and home ownership through the tax system, one could envision a simpler, more efficient, and fairer approach than the current mix and design of tax expenditures. One of the benefits of the consolidation approach is that the Department of Fi- nance already organizes its tax expenditures along thematic lines in its annual report on federal tax expenditures. A multi-year review process could be organized on the basis of the report’s themes or a subset of them. Another benefit is that the consolidation approach aligns with the three political economy conditions for successful reforms. It is relatively simple, since it is limited to a single policy theme; and because it targets an identifiable set of taxpayers, it helps with communications and designing a distributionally neutral package. It also prevents a political lag or policy uncertainty between the elimination of any tax expenditures and the related announcement of new policies. How would this approach work in practice? Each year the government could tackle some number of thematic groups with the goal of simplification and greater progressivity in the personal income tax sys- tem, and simplification and efficiency in the corporate income tax system. The exercise could target revenue neutrality, aim for net revenue gains as part of an overall fiscal strategy, or even incur net costs if the reforms were part of a broader policy strategy, such as the caregiving example. The reviews could be conducted each summer, and the results could be effectuated in the subsequent budget. The review process would be managed by the Department of Finance, but it would necessarily involve input and perspective from officials in other departments as well as external experts. The involvement of departmental officials would be -im portant to understanding the interaction of tax-based policies with other, related federal programs and the relative utility of those policies. Think, for instance, of the role of tax expenditures related to home ownership and the Canada Mortgage and
40 These include the first-time home buyers’ tax credit, the non-taxation of capital gains on principal residences, the home buyer’s plan, the GST exemption for certain residential rent, the GST rebate for new housing, and a rebate for new residential property. Canada, Department of Finance, Report on Federal Tax Expenditures—Concepts, Estimates and Evaluations 2019 (Ottawa: Department of Finance, 2019). The list could also include the non-taxation of imputed rent, which is not classified as a tax expenditure by the department but is considered by some economists as a deviation from a pure Haig-Simons tax base. 46 n canadian tax journal / revue fiscale canadienne (2020) 68:1
Housing Corporation’s mortgage insurance. It would also be useful to draw on analyses by external experts to determine the effectiveness of different tax expendi- tures. Think, for instance, of academic research on the children’s fitness tax credit41 or the labour-sponsored venture capital tax credit.42 In order to inform and guide the thematic reviews, the government could make use of a standardized checklist to evaluate tax expenditures. This would help to inform decisions about possible reforms similar to the tests applied to program spending in the strategic review process. The US Government Accountability Office (GAO) has produced a useful checklist for evaluating tax expenditures.43 The framework sets out criteria and analytical questions for policy makers to consider in weighing competing priorities and evaluating the merits or effectiveness of a particular tax expenditure. The checklist consists of five considerations, each involving a series of tests or metrics:
1. What is the tax expenditure’s purpose, and is it being achieved? 2. Even if its purpose is being achieved, is the tax expenditure good policy? 3. How does the tax expenditure relate to other federal programs? 4. What are the consequences of the tax expenditure for the federal budget? 5. How should evaluation of the tax expenditure be managed?
There is room to improve the GAO’s framework. First, item 2 should be changed to add “Is the tax expenditure the best way to achieve this goal?” There may be worthy goals that are not best supported or promoted through the tax system. Second, item 4 should be changed to consider the interaction and effect of federal changes on provincial responsibilities such as health care and education. The government could use the checklist for internal purposes only, or it could release the results as part of its annual budget, where it would ostensibly enact any resulting policy changes. Making the completed checklists public could have pol- itical economy benefits by clearly and comprehensively outlining the justifications for reforms in a standardized form. And it would require the inevitable critics to contend with the substantive results of the review rather than merely criticize the government’s motives.
41 John Spence, Nicholas Holt, Julia Dutove, and Valerie Carson, “Uptake and Effectiveness of the Children’s Fitness Tax Credit in Canada: The Rich Get Richer” (2010) 10 BMC Public Health (https://doi.org/10.1186/1471 - 2458 - 10 - 356). 42 Douglas Cumming, Jeffrey MacIntosh, and Keith Godin, “Crowding Out Private Equity: Canadian Evidence,” Fraser Alert, September 2007 (https://pdfs.semanticscholar.org/bab0/ 190103eae482746f6c359872024cdf4f247b.pdf ). 43 United States, Government Accountability Office,Tax Expenditures: Background and Evaluation Criteria and Questions, document no. GAO- 13 - 167SP (Washington, DC: GAO, November 2012) (www.gao.gov/products/GAO- 13 - 167SP). policy forum: tax reform in canada’s 43rd parliament n 47
The use of such a checklist would not remove the role for preferences and values, but it would root our normative debates in a common set of evidence and facts. This would mitigate one of the biggest challenges arising from the small-business tax controversy in 2017. Normative differences in that case were exacerbated by unclear and competing understandings of the facts.44 Moving through the tax system on an incremental and thematic basis may not produce fundamental change, especially in the short term. But if the government were able to replicate the caregiving-related reforms across the tax system, we would, over time, incrementally get a simpler, more efficient, and fairer tax code. That strikes me as a highly satisfactory second-best outcome.
CONCLUSION As the federal government determines how to deliver on its commitment to review the federal tax system, it should resist calls for a royal commission or other ap- proaches that would attempt to divorce tax policy from politics. Instead it should pursue a model that leans into politics and political economy insights about the conditions for successful tax reform. An incremental yet systematic model based on the Harper government’s stra- tegic review process can help the government to make progress on rationalizing the federal tax system. It will be less ambitious than some stakeholders would prefer and may require second-best choices. But this is unavoidable in an environment with a multiplicity of normative perspectives on the goals and design of tax policy. The reform process will necessarily require moderation and compromise. And that requires politics. Still, if such a process for policy review and reform is well structured and prop- erly executed, it can achieve steady and ongoing progress in the direction of ration- alization of the federal tax system. This ought to be a basis for policy and political convergence in the context of a minority Parliament and into the future.
44 Sean Speer, “Who’s Right in the Small Business Tax Controversy?” Macdonald-Laurier Institute Inside Policy, September 15, 2017 (www.macdonaldlaurier.ca/whos-right-small-business-tax -controversy-sean-speer-inside-policy). canadian tax journal / revue fiscale canadienne (2020) 68:1, 4 9 - 6 7 https://doi.org/10.32721/ctj.2020.68.1.pf.gillezeau
Policy Forum: Expenditures, Efficiency, and Distribution—Advice for Canada’s 43rd Parliament
Rob Gillezeau and Trevor Tombe*
PRÉCIS Cet article examine les engagements de dépenses, individuellement et globalement, qui ont été pris par le Parti libéral du Canada lors de la 43e élection fédérale canadienne en octobre 2019. Les auteurs indiquent quelles politiques proposées sont relativement efficaces dans leur conception ex ante, quelles politiques pourraient être modifiées pour améliorer les objectifs d’efficacité ou de répartition, et quelles politiques ont des bases politiques limitées. En général, ils soutiennent que le gouvernement devrait adopter une approche ciblée, en choisissant bien les mesures et en limitant le nombre de nouvelles initiatives simultanées. Ils concluent en indiquant quels domaines pourraient faire l’objet d’une collaboration avec les trois autres grands partis du 43e Parlement, en particulier les possibilités d’améliorer les régimes de transferts aux provinces.
ABSTRACT This article examines the expenditure commitments, individually and in aggregate, from the Liberal Party of Canada in the 43rd Canadian federal election in October 2019. The authors articulate which proposed policies are relatively efficient in their design ex ante, which could be adjusted to improve efficiency or distributional goals, and which have limited policy grounding. In general, they argue that government should take a focused approach to expenditures, both with respect to appropriately targeting measures and in limiting the number of simultaneous new initiatives. They conclude by indicating areas of potential cooperation with the other three major parties in the 43rd Parliament, with a particular emphasis on possibilities to improve provincial transfer regimes. KEYWORDS: EXPENDITURES n FISCAL PLANNING n TRANSFERS n REVIEWS
* Rob Gillezeau is of the Department of Economics, University of Victoria (e-mail: [email protected]). Trevor Tombe is of the Department of Economics and the School of Public Policy, University of Calgary (e-mail: [email protected]).
49 50 n canadian tax journal / revue fiscale canadienne (2020) 68:1
CONTENTS Introduction 50 A Critical Perspective on LPC Policy Commitments 52 Aggregate Fiscal Track 52 Individual Transfers 55 Health Care 57 The Environment and Climate Change 58 Child Care 60 Housing 60 Other Expenditure Measures 61 Minority Party Expenditure Commitments 62 New Democratic Party 62 Bloc Québécois 63 Conservative Party of Canada 64 Recommendations for the New Parliament 64 Expenditures and Individual Transfers 65 Provincial Transfers 65 Efficient Public Spending 66
INTRODUCTION The 43rd federal general election in October 2019 left Canada with a fractured minority Parliament, in which the governing party—the Liberal Party of Canada (LPC)—was returned with the smallest share of the popular vote in Canadian hist- ory.1 Despite this, Prime Minister Justin Trudeau has indicated that he will govern without a formal or informal arrangement with any of the opposition parties;2 essentially, the government will operate in a manner similar to its majority pre- decessor, given the likelihood that it can find a supportive partner for most potential policies.3 And while governing involves choices that go beyond platform commit- ments—which will be unavoidable as unexpected events unfold or administrative and implementation challenges appear—exploring those commitments is valuable.
1 The 43rd Parliament is unique in its divisions with respect to the popular vote. Prior to the 33.07 percent received by the LPC in the 2019 election, the lowest share of the popular vote for the governing party was 35.9 percent for Joe Clark’s Progressive Conservatives in 1979. The first- and second-place finishers in 2019, in total, also received a historically small share of the vote. See Parliament of Canada, “Elections and Ridings” (https://lop.parl.ca/sites/ ParlInfo/default/en_CA/ElectionsRidings). 2 Kathleen Harris, “Trudeau Rules Out Coalition, Promises Gender Equity in New Cabinet,” CBC News, October 23, 2019 (www.cbc.ca/news/politics/trudeau-liberal-minority-government -2019-1.5331926). 3 With each of the Conservative Party of Canada (CPC), the New Democratic Party (NDP), and the Bloc Québécois (BQ) holding sufficient seats to independently ensure the survival of the government, there is a broad range of policy spaces in which they could feasibly operate while maintaining support. policy forum: expenditures, efficiency, and distribution n 51
To that end, we consider the policy agenda that the LPC offered in the 2019 election and offer suggestions as to how the values articulated by the governing party, and supported by Canadian voters across the political spectrum, may be most effectively and efficiently realized. This is a particularly useful task because electoral platforms, even with their increasing technical complexity, largely remain signals of values rather than providing specific policy prescriptions.4 We begin with an overview of the spending path proposed by the new govern- ment and provide important historical context. We then consider major spending proposals across a range of areas, relating them to the broader economic literature and the historical context, and suggest how they may be improved. Given the size and scope of individual transfer programs, we start with the proposals in this area laid out during the campaign. Fully one-third of promised spending increases are found in an enlarged old age security (OAS) benefit and Canada child benefitCCB ( )— specifically, a 10 percent increase to OAS for individuals over the age of 75 and a 15 percent increase to the CCB for parents of newborns. We argue that while most of the proposed transfers are reasonable, addressing poverty for those over 75 through OAS is not ideal. Rather, much more could be done through targeted measures, such as a substantially expanded guaranteed income supplement (GIS), for the same ag- gregate cost. We then move on to exploring a number of direct program expenditure propos- als, including health care, the environment and climate change, housing, child care, and other measures. While there are many important details yet to be determined, we highlight several concerns and principles, including the desirability of focusing resources on a small number of programs, the need for piloting otherwise untested policy measures, and—an overarching consideration relevant for many of these pro- grams—appropriate federal-provincial cooperation given the jurisdictional-fiscal divide. Regarding the latter, we argue that it is reasonable for the federal govern- ment to be involved if there are potential spillovers from an area of public spending, if there are economies of scale that cannot be fully realized at the provincial level, if harmonization enhances national-level efficiency, or if an area is already one of joint federal-provincial jurisdiction in practice.5 Finally, we extend this discussion to areas of viable compromise with potential minority partners, spanning a range of fiscal options, and conclude with a set of selected recommendations for the new government. We argue that the government should be highly focused in its decisions on new expenditures or revenue reductions,
4 This is not to understate the scale of the transformation to platforms with more policy content; most major parties now include viable fiscal plans in their campaign materials. Further, the introduction of high-quality costing services through the Office of the Parliamentary Budget Officer (PBO) should further the shift from values articulation to concrete, well-defined policy proposals. 5 While not necessarily ideal, federal interventions in growth-enhancing areas that may help to relieve the federal-provincial fiscal imbalance are also reasonable in our opinion. 52 n canadian tax journal / revue fiscale canadienne (2020) 68:1 choosing a focus on “affordability,” service provision, or strengthening the fiscal position of the federation. We argue that it is particularly important for the new government to begin to address the large and growing fiscal challenges of provincial governments.6 The federal government can help to mitigate some of this pressure by increasing the growth in health-related transfers, for example. Another option— which may be adopted in lieu of or as a complement to expanded health transfers— is direct federal spending on various health-care measures; in particular, multiple parties and the federal government have clearly signalled their commitment to pharmacare and, potentially, government-funded national dental care. We conclude by noting the appropriateness of a program review to provide additional fiscal space to meet priorities and offer some recommendations as to its approach.
A CRITICAL PERSPECTIVE ON LPC POLICY COMMITMENTS The 2019 election campaign marked a shift in values articulated by Justin Trudeau’s Liberal Party. In the 2015 campaign, Mr. Trudeau set out an agenda grounded in an expanded state with a substantial increase in means-tested transfers to individuals.7 The 2019 campaign saw a shift from expenditure growth to tax reductions as the core policy offer. Even with this shift, there are substantial expenditure commit- ments in the LPC’s platform that, in aggregate, are larger than the proposed tax reductions. Such proposals should be evaluated against best practices and empirical evidence.
Aggregate Fiscal Track The LPC’s commitments include moderate revenue and expenditure changes that we detail below. With respect to revenues, the cost of the LPC proposal to increase the basic per- sonal amount totals nearly $5.7 billion per year in forgone revenue by 2023 - 24. This is partially offset by proposed new revenues that include a 10 percent luxury goods tax, a national tax on vacant properties, a proposed “crackdown on corporate tax loopholes,’’ a levy on certain multinational technology companies, and other
6 The PBO has repeatedly noted that the long-run challenges facing provinces are substantial. An aging population in particular will strain health-care budgets for decades. See, for example, Office of the Parliamentary Budget Officer,Fiscal Sustainability Report 2018 (Ottawa: PBO, September 2018). 7 Principal among these measures was the introduction of the CCB. For a thorough review of the transformation of child benefits and its impacts, see Adriene Harding, The Effect of Government Transfer Programs on Low-Income Rates: A Gender-Based Analysis, 1995 to 2016, Income Research Paper series, catalogue no. 75F0002M (Ottawa: Statistics Canada, 2018). policy forum: expenditures, efficiency, and distribution n 53 measures.8 Overall revenue increases amount to $3.7 billion by 2023 - 24. We illus- trate the aggregate changes, as a share of gross domestic product (GDP), in figure 1.9 Turning to expenditures, the LPC proposed increased government operations and transfers. While not all commitments may be cleanly separated between direct operating spending by government, transfers to individuals, or transfers to govern- ments and other entities, the majority of LPC spending commitments are increased transfers to individuals.10 We explore individual proposals below. In aggregate, expenditure commitments exceed $11.3 billion per year by 2023 - 24. Over the next four years, this represents an average increase of 0.4 percent of GDP, but is partially offset by spending reductions through a “tax expenditure and government spending review.”11 The average net spending increase is roughly 0.3 percent of GDP.12 With respect to direct program spending, the LPC platform implies a decrease from the current level of 14.7 percent of GDP to 14.3 percent by 2023 - 24. We illustrate this change in figure 2, comparing the spending path articulated in the LPC platform with the projected fiscal path generated by the Office of the Parliamentary Budget OfficerPBO ( ) prior to the election. From a macroeconomic perspective, the change in the federal fiscal stance pro- posed by the LPC is modest.13 Importantly, the government retains the fiscal space
8 Liberal Party of Canada, Forward: A Real Plan for the Middle Class (Ottawa: LPC, 2019), at 79 - 80 (https://2019.liberal.ca/wp-content/uploads/sites/292/2019/09/Forward-A-real -plan-for-the-middle-class.pdf ). 9 We compare changes in revenues and expenditures under the LPC platform with the projected fiscal path generated by the PBO prior to the election commitments. In this analysis, we exclude the incremental revenue attributed to the trans-mountain expansion project, which reflects estimated corporate tax revenues (largely in upstream oil and gas) from the project. In our view, the baseline projections implicitly incorporate such revenues, and therefore the amount reported in the LPC platform is not incremental. 10 For example, boosting OAS benefits by 10 percent for individuals over the age of 75, increasing the CCB for the first year of a child’s life, doubling the Canada child disability benefit, increasing Canada student grants, and so forth. 11 Forward: A Real Plan for the Middle Class, supra note 8, at 79 - 80.There is no indication of what share of this review will draw from operating expenses versus tax expenditures. As we discuss later, a program review focused on new expenditures since 2015 should be able to recover a reasonable share of these allocated dollars. 12 For perspective, this will maintain the overall size of federal expenditures at roughly 15.6 percent of GDP at the end of the four years, fully offsetting the decline to less than 15.3 percent under the PBO’s baseline projection. 13 The implied change in the federal debt-to-GDP ratio as a result of the LPC platform commitments is also modest. This metric is central to long-run debt sustainability and therefore an important gauge of a government’s overall fiscal stance. Previously accumulated federal debt grows at the rate of interest while the aggregate federal tax base grows with GDP. So long as growth rates exceed interest rates, the federal government can sustain a primary budget deficit—the difference between spending and revenues without considering debt costs—in perpetuity. And under the LPC platform commitments, the federal primary balance is positive, with a projected surplus of roughly $13 billion in 2023 - 24. The sustainability of the 54 n canadian tax journal / revue fiscale canadienne (2020) 68:1
FIGURE 1 Change in the Projected Federal Budget Deficit, 2019-20 to 2023-24, Under the 2019 LPC Platform 1.2
1.0
0.8
0.6
0.4 Share of DP (%)
0.2
0.0 PBO pro ected New Tax New Spending LPC pro ected budget deficit revenues reductions spending reductions budget deficit