White Knights or Machiavellians? Understanding the motivation for reverse in Singapore and Thailand

Pantisa Pavabutr1

Thammasat University

This draft: October 2017

Abstract

We analyze the characteristics, return, and financial performance of 47 RTO sample firms in Singapore and Thailand between 2007‐2015. Given the existing regulatory screens imposed by Singapore and Thai exchanges, we cannot find evidence that firms choosing to list via RTOs signals a separating equilibrium of low type firms seeking a listing short‐cut. We argue that RTOs should not be evaluated as a choice of listing per se, but as part of the parcel of long‐term corporate strategy. Analysis in this paper yields important in sights on reverse takeovers from both investor and regulator perspectives.

JEL Classification: G14, G34.

Keywords: Reverse takeovers, Back‐door listings, Emerging markets.

1. Introduction

Equity markets have both an allocation and monitoring role. The challenge for regulators is striking a delicate balance between overseeing transparency and fair rules governing listings for efficient allocation of resources and investor protection without delineating potential firms from entering organized exchanges. The dual paths to listing can be direct through an initial (IPO) or indirect through a reversed (RTO). An IPO is traditionally seen as a young company with full listing qualification coming of age and offering shares to the public. A reversed takeover is the process whereby a private company acquires a controlling stake in a in order to obtain listing status. In doing so, RTOs allows private firms to list and seek out growth opportunities by merging with public firms and to seek listing without too much dilution and vulnerability to market conditions. However, anecdotal and some selected empirical evidence suggests a dark side to RTOs as they are often referred to as “back‐door listings” with transactions associated with opaque firms wanting to bypass stringent listing rules or a method in which is a holding company of the shell firm tries to get rid of non‐performing assets by passing it along to the next uninformed investor.

Under this setting, there are two issues where reversed takeovers are subject to debate. First, should reverse takeovers be allowed or should regulations be tightened? The second issue, is why should firms choose RTOs instead of IPOs? The drawbacks of RTOs is that their speed and cost saving benefits are often overestimated in particular with on‐going trends towards more regulatory scrutiny

1 Associate Professor of Finance, Thammasat Business School, [email protected]. The author wishes to thank the National Research Council, and Thailand Research Fund for grant support. The author appreciates research assistance from Sippapon Ithiprachayaboon, Prasit Sutthikamolsakul, and Nina Taphunwong. Comments welcomed.

1 pertaining RTOs in equity markets around the world (Sjostrom, 2008; Vermeulen , 2014). Pavkov (2006) suggests that a complete analysis of benefits and costs of RTOs should include all stakeholders involved.

Our study contributes to the almost non‐existing research on RTOs in ASEAN. Using 47 RTO sample firms in Singapore and Thailand between 2007‐2015, this paper addresses the above questions through discussion of existing listing regulations on RTOs and empirical analysis return and financial accounting performance. More specifically, it first briefly reviews existing regulations in Singapore and Thailand whether they are conducive for firms to list by RTOs rather than by IPOs. Second, it empirically analyzes the RTO samples by posing these questions 1), What are the characteristics of RTOs transactions? 2) What is the investors’ experience in RTO transactions over short and long‐term periods, and 3) What is the financial accounting performance of merged entity.

The paper finds that given the existing regulatory screens imposed by Singapore and Thai exchanges, we cannot find evidence that firms choosing to list via RTOs signals a separating equilibrium of low type firms seeking a short‐cut to listing. This is because both exchanges require the newly merged entity to file a reapplication and comply with the same minimum standards as IPO listings. From company circulars and financial advisors report of sample firms, the number of days to completing the RTO transaction can be up to one year. In fact, about half of the RTO samples involve listing on the main boards of Singapore Exchange (SGX) and Stock Exchange of Thailand (SET); the other half on the second boards of Catalist (CAT), and Market for Alternative Investment (MAI). If a RTO is a venue for easy listing, then surely more transactions would have occurred on secondary boards where listing requirement is more flexible, in particular on CAT. Second, after examining the characteristics of sample RTO deals, we find 32 firms or 68% of the sample firms are classified by the auditor as financially distressed, and approximately half of the transactions occur across firms with different industries. This implies that outsider firms intend to takeover a “shell” or non‐performing firm in an unrelated industry. Next, we examine the investor’s experience by measuring short‐term abnormal returns around RTO announcements and long‐term 12 months return post‐announcement. The cumulative abnormal return (CAR) of distressed firm RTOs during event window [‐10,10] around RTO announcement (MOU date), is 28.8% significantly higher than 8.4% of non‐distressed firms. Over a holding period of 12 months in which the buy‐and‐hold abnormal return (BHAR) is evaluated, much of the original short‐term gain reverted primarily because of continuing capital increases. Even so, the mean BHAR of RTO samples are higher than the mean BHAR of control firm sample. Looking at financial performance, distressed firms’ net profit margins and return on assets improved after RTO completion, but non‐distressed firm performance remains relatively superior to distressed ones.

In sum, in our analysis of short‐term and long term return performance we do not find evidence that our sample RTOs transactions are causing market instability through fraud and pump‐and‐dump schemes. If the ease of listing is not a reason for choosing RTO, then why do firms engage in these transactions? From our analysis, we view that RTOs should not be perceived as a listing tool alone but part of the parcel of long‐term corporate strategies where (i) the private firm can become a public company without immediate large dilution in ownership and still have an option to raise more funds later (many RTO transactions have issue plans contingent on successful restructuring plans post‐transaction); (ii) the private firm wishes to takeover a distressed or non‐distressed target public firm by obtaining a bargain price since the target is substantially smaller and trading at a value weighted average price (VWAP) in the bottom decile trading range of the exchanges; (iii) the public firm controlling shareholders wishes to exit and is offered an attractive price to the private firm. RTO

2 transactions do provide a more liquid trading environment which facilitates exit, and iv), the motivation could be unique to the firms themselves. 2

There are altogether five sections in this paper. Section 2 provides background discussions on RTOs including deal structure, regulations, and related literature. Section 3 describes the sample data and describes key features of RTO transactions. In section 4, both descriptive and empirical methods are used to analyze the motivation for RTOs. Finally, section 5 concludes and discuss policy implications for capital market development.

2. Background on RTOs

2.1 Deal structure

In a typical RTO the private company acquires a controlling stake in a public company in order to obtain listing status by allowing the listed firm to acquire its assets or equity and in return receive issued shares of the listed company.

Figure 1: Illustration of reverse takeover

Before transaction

Shareholders of Shareholders of listed firm private firm

Cash transfer, or share swap

Listed firm Private firm

Transfer assets After transaction

Shareholders of Shareholders of private firm listed firm (with more than 50% control)

Merged firm

2 We also expect that by acquiring a loss‐making public firm, the private firm gains from . However, we do not have unconsolidated financial statements of private firms to evaluate this conjecture. In Singapore’s case, some foreign outsider firms are already listed overseas, but explain that they prefer to list on the Singapore exchanges to increase their international profile in Asian markets. With RTOs they do not need to pay fees. Other outsider firms conduct RTOs as a short‐cut to international diversification (Perrenial China Retail Trust transaction with St. James Holding) or a short‐cut to business diversification (Singha Estate transaction with Rasa Property on SET).

3

Figure 1 illustrates the reverse takeover process. In exchange for assets transfer of private firm, the listed firm may choose from one or combination of payment in cash, new shares issue through private placement, borrow, or issue new shares to swap with private firm. In a RTO, the listed firm is often referred to as a shell firm if it has the following properties: (a) no or nominal operations; and (b) either: (1) no or nominal assets; (2) assets consisting solely of cash and cash equivalents; or (3) assets consisting of any amount of cash and cash equivalents and nominal other assets.

Not all firms in our Singapore and Thai samples start out as “shell” firms; however, in the reverse takeover process they can become one by liquidating their businesses or by agreeing to spin off its current assets and liabilities at the time of the reverse merger. Some listed firms are profitable but are struggling with poor business growth opportunities and seek reverse merger partners to strengthen their industry positions through diversification and synergy.

Figure 2: Controlling structure in reverse take‐overs Value $ No of mn shares % Own Private firm shareholders 10 50 79% Private Private placement 2 10 16% placement $ Public firm 2 mn (10 mn shareholders 0.6 3 5% shares x $0.2) 12.6 63 100%

Merged firm

Combined value $12.6 mn (63 mn shares x $0.2

1:1 2:1

Private firm Public firm value $ 10 mn value $ 0.6 (50 mn shares mn (3 mn new x $0.2) shares x $0.2)

Figure 2 shows a reverse take‐over between a public shell firm valued at $600,000 with share priced at $0.1 and 6 million old shares outstanding and a private firm valued at $10 mn with 50 mn shares outstanding at $0.2 per share. The financial advisor expects the merged entity to require $2mn addition funds for restructuring process and issues 10 mn shares at $0.2 as private placement. Assume for simplicity the combined firm value is estimated to remain at $0.2 per share (post‐consolidation issue price). The public shell shareholders are offered pre‐consolidation issue price of $0.1 per share and then

4 given share swap at a rate of 2 to 1 such deriving a post‐consolidated issue price3 of $0.2 per share and that their original ownership value is set at $600,000. As a result, the new combined value of the firm is $12.6 mn with private firm owners accounting for 79% (50/63) of merged firm control whereas private investors account for 16% (10/63) and public shell shareholders account for 5% (3/63).

2.2 Reverse takeover regulations and process

An RTO is perceived as a “short‐cut” for a private (outsider) firm to achieve listing status and bypass the lengthy process of an IPO and hence often referred to as “backdoor listing.” We discuss the legal of definition of RTO and compare and contrast the IPO and RTO process next.

Both Singapore and Thai Exchanges apply bright line tests for two specific types of RTOs which are transactions involving a change in control of a listed issuer and a very substantial acquisition (VSA).4

Table 1 IPOs vs RTOs: Process

IPO RTO (1) Prelisting restructuring and due diligence of (1) Negotiation and due diligence between the firm in order to comply to listing criteria and listed firm and the private (outsider) firm leading ready firm for public disclosure. to an MOU or sale and purchase agreement (SPA). (2) Preparation of prospectus and application (2) Preparation of circulars to shareholders and submission to SEC and SET. The prospectus for stock exchange approval. Circulars contain contains disclosures required regarding business description of the transactions, financial and firm. information of target group and merged group. (3) Public exposure: Road shows, nomination of (3) Obtain approval from extraordinary underwriter, and share subscription and shareholder meeting (EGM). distribution. (4) Trading commences (4) Disposal of assets of listed firms (if necessary) and mandatory . Acquisition completed and trading of merged group begins.

Table 1 summarizes the IPO and RTO process. It is apparent that a RTO process bears many similarities with an IPO process. Both begins with the need for detailed due diligence and preparation of prospectus or circular containing disclosures regarding the deal, compliance with the usual listing criteria, and final approval from the exchanges.5 All in all both process is likely to lead to similar

3 Sometimes, the issues are further consolidated at a set ratio by the financial advisor to reduce the impact of dilution or to comply minimum listed stock price rule. The minimum trading price rule on Singapore exchange is to maintain past 6 months VWAP at no less than S$ 0.2 thus, consolidation of new share value must meet this minimum requirement. 4 These are a listed issuer's acquisition (or series of acquisitions) of assets where any percentage ratio is 100% or above in terms of net tangible assets, net profit, total consideration, equity value, or proven and probable reserves. Details are available from SGX rulebook Chapter 10, Part VIII section 1015 and SET’s listing rule 11‐00. In markets that are more relaxed, for example Canada, and the US, a transaction is regarded as an RTO only when control of incumbent shareholders  50% post‐transaction. Winyuhuttakit (2011) gives excellent details and analysis on RTO regulations in Hong Kong, Korea, Thailand, and US. 5 A waiver for listing reapplication may be requested if a RTO involves same or similar industry transactions, maintains the same core business of the listed incumbent firm, and no change in board and control.

5 timeframe to completion. However, dating the length of time required for the RTO process is not straightforward as Table 1 may suggest.

A RTO transaction can be complicated further with more parties involved. As a matter of fact, it begins with negotiation and due diligence process both public and private firms potentially adding time and cost to the transaction and represents and “double‐lemons” problem. Once the terms and price are agreed, upon board approval, a memorandum of understanding (MOU) is signed. The terms in the MOU are disclosed to the exchange for reapplication of the merged entity for listing. Then after the parties received the go ahead from the exchange, an extraordinary shareholder meeting (EGM) is called requiring a super majority support. The merger can then legally begin as consolidated trading of shares can start and hence the shares of the new entity can take on new value from that point.

With the process described above, we can say that there are two important event dates in the process, the MOU date, which contains most information impact, and the EGM date that marks the beginning of consolidated trading of shares of the merged entity. However, the true completion of the acquisition can be lengthened as the merged firm may continue to carry separate share placement exercise to meet capital needs or listing requirements on free float.

2.3 Related literature

RTOs have been used as an alternative means to list on an exchange for decades. Historically, no significant regulatory review was required resulting in a shorter timeframe for listing completion and substantial costs saved in terms of underwriting fees as the process requires neither a prospectus nor an underwriter. In addition, timing of RTOs are not subject to market conditions as in IPOs. New controlling owners of the firm generally suffer less share dilution, obtain the public firm at a relatively economic price as opposed to regular mergers as well as an option to raise funds in the future. Incumbent shareholders of the public firm benefit from a clearly defined exit strategy through the public market.

Given a history of flexible regulatory environment, early research on RTOs tend to suggest a separating equilibrium where high quality firms choose listing via IPOs and low quality firms choose listing via RTOs. In Arellano‐Ostoa and Brusco (2002), a high type firm, distinguished by greater probability of obtaining positive NPV projects, choose to signal with IPO whereas low quality firm with low probability of positive NPV projects choose RTOs. The necessary conditions for a separating equilibrium in the model is that the expected NPV of the high type firm is greater than that of the low type firm and that the cost of completing IPO is higher than completing RTOs. A number of empirical papers mainly based on US evidence support a separating equilibrium in listing choice as they find firms choosing to list via RTOs instead of regular IPOs tend to be smaller firms with relatively higher level of information asymmetry. Gleason et al. (2005) studies 121 RTO cases between 1987‐2001 AMEX, NYSE, NASDAQ and find that they tend to be speculative in nature and fail to generate long‐term wealth gains. In their sample, 46% RTO firms survived after first two years of listing. Using a slightly different period on US market between 1990‐2002, Adjei, Cyree, and Walker (2005) document 42% of their sample firms become delisted in first three years. Floros and Shastri (2009) find that firms choosing to list via RTOs instead of regular IPOs tend to be smaller firms with relatively higher level of information asymmetry. Thus, the authors view that RTOs cannot be compared to regular IPOs but rather to penny IPOs. Floros and Sapp (2011) focuses on of shell firms traded on OTC or pink‐sheets that go through RTOs. They find up to 48% in abnormal return in RTO announcements, but the long term performance erases the gains as surviving firms earn post‐even annual return of ‐91%. The Canadian experience

6 with RTOs is similar as Carpentier, Cumming, and Suret (2009) documented lower quality firms opt for the less regulated RTOs to obtain public listing . Wan‐Hussin (2002) documents the case of high profile reverse take‐over completed in 1995 in Malaysia that resulted in a backdoor listing of a private company, Jaya Tiasa Plywood, via listed Berjaya Textiles. The author finds increase wealth effects of both minority shareholders and the former controlling shareholders. However, the work is limited to one particular case and only short‐term wealth effects were documented.

Absence of early regulatory oversight has led to abuses and fraudulent conduct leading to tightened RTO regulations. In US, increased reports of fraudulent activities in late 2000s has prompted SEC to apply more stringent listing rules to list publicly. These additional requirements include minimum share price maintenance, complying to filing requirements of financial reports, and a seasoning rule that requires the merged entity to trade on OTC prior to official listing. Following fraudulent cases of Chinese firms backdoor listing, the Hong Kong Exchange introduce the Main Board RTO Rules in March 2004 which virtually eliminated the practice of injecting non‐listed assets without a suitable track record for listing into a listed shell in conjunction with a change of control. 6

Is the verdict out on RTOs? A look at RTO critique from legal and regulatory point of view offer other perspectives. Pavkov (2006) encourages more empirical analysis of RTOs to help settle the policy debate whether more regulatory intervention is necessary. Vermeulen (2014), notes that in recent years, there is a dramatic increase in RTOs on ASEAN exchanges. Existing work on RTOs are primarily limited to the US experience and cannot be comparable to the ASEAN experience where demand‐supply and regulatory environments are very different. For example, the Canada and the US applied a laissez‐ faire approach towards RTOs in the past in order to support small and micro‐cap companies who were unlikely to be able to afford the underwriter necessary for an IPO.

Singapore and Thailand on the other hand adopt a more conservative approach towards regulating such that firms that undergo RTOs must undergo reapplication process to list on the exchange and must meet the same minimum listing requirement as IPO firms. This means that the cost of conducting RTOs in Singapore and Thailand is likely to be quite high. Although the direct costs of underwriting fees present in regular IPOs is absent in RTOs, private firms engaging in such transactions face numerous indirect costs. For example, the risk of overpaying for the shell firm if the financial advisors’ place the value of the shell firm too high. There is also cost of restructuring the shell firm and the risk that synergy value of the merged entity in a vertical or horizontal merger within firms of the same industry does not materialize. Sjotrom (2008) argues that the costs of RTOs is underestimated, after factoring in indirect costs, he finds that actual costs of IPOs and RTOs are in fact not that much different. In the case of Singapore and Thailand, the reapplication requirement and disclosure effort means that the speed advantage is unlikely to be that great. Thus, if the regulatory costs and indirect costs are high, then why do firms still choose to go public with RTOs? Hsieh, Lyandres, and Zhdanov (2009) propose a model that link’s a firm’s decision to go public with subsequent takeover strategy. Completing an IPO reduces valuation uncertainty leading to a more efficient acquisition strategy.

6 See Rules on Backdoor Listings: a Global Survey, OECD, December 2014 and Charltons Solicitors, www.chartonslaw.com.

7

3. Sample Data Analysis

Between 2007‐2015, we identify RTO cases from stock exchange and SEC websites. A list of Singapore RTO cases can be found of SGX website under the “Catalodge” page where important firm disclosure information is publicized. Thai RTO cases can be tracked from SEC’s website where tender offers, material asset acquisitions and disposals of listed firms are reported. 7 Details of individual RTO deals are gathered from a combination of financial advisor reports and minutes of shareholders meeting. We collect a total of 42 RTO cases in Singapore and 17 cases in Thailand. We find that ten RTO transactions of the 42 in Singapore were aborted whereas two transactions of 17 were unsuccessful in Thailand. The search is biased towards successful cases as information is more readily accessible.

Data is collected from various sources. We obtain market trading data from Datastream and listed company accounting information from Worldscope. Our sample includes firms listed on both the main and secondary boards in which case it is the Singapore Exchange (SGX) and Catalist (CAT) for the Singapore sample and Stock Exchange of Thailand (SET) and Market for Alternative Investment (MAI) for the Thai sample.

Table 2 Distribution of RTO activities 2007‐2015

The RTO sample reports the distribution of RTO activities announced by Singapore and Thailand exchanges. This table report the RTO activities and the proportion of transactions along side with SGX and SET year end index level and annual return. The sample includes firms on the main boards (SGX and SET) and secondary boards (CAT and MAI).

No of firms % of Market index Market return Year sample All Singapore Thailand All Singapore Thailand Singapore Thailand

2007 1 0 1 2% 3,466 858 19% 26%

2008 1 1 0 2% 1,762 450 ‐49% ‐48%

2009 3 1 2 5% 2,898 735 64% 63%

2010 10 8 2 17% 3,190 1,033 10% 41%

2011 6 5 1 10% 2,646 1,025 ‐17% ‐1%

2012 10 8 2 17% 3,167 1,392 20% 36%

2013 16 13 3 27% 3,167 1,299 0% ‐7%

2014 10 4 6 17% 3,365 1,498 6% 15%

2015 2 2 0 3% 2,883 1,288 ‐14% ‐14% Total 59 42 17 100%

7 http://infopub.sgx.com and https://www.set.or.th/set publishes Singaporean and Thai listed firms’ important announcements and circulars. Day of MOU of VSA appears here, where MOU date is unavailable we use date when first circular is publicly available from firms’ website or news sources.

8

Table 2 presents descriptive statistics for the sample RTO transactions in Singapore and Thailand by sample year. The number of RTO activities for Singapore is highest in 2013 with 13 transactions whereas for Thailand it is highest in 2014 with six transactions. The last four columns of Table 2 provide end of year market index level and annual market return. There seems to be no apparent association between market performance and intensity of RTO activities in Singapore. For example, in 2009 when the annual market return on SGX surged by 64% only one RTO transactions came through. In the following years 2010‐2011 when the market rose 10% and dropped 17% the number of transactions was eight and five respectively. In Thailand’s case, RTO transactions appear to pick up when market performance improves. As shown in 2014, the SET gained 15% following a 7% loss in 2013. The number of reported transactions increased from three to six between 2013‐2014.8

Table 3 reports successful deal firm statistics which are segmented by country of listing and then by distress or non‐distress status. Where relevant we report both average values and median values in parentheses below. There are a total of 47 successful cases of which 32 are listings on Singapore exchange and 15 on the Thai exchange. Approximately half of the entire sample firms (51%) are declared financially distressed by the listed firm’s auditors. Data of the two markets combined shows that RTO transactions are quite evenly split between main and secondary boards and between transactions that occur within the same industry and across industries. Singapore sample firms are slightly in favor of secondary board transactions (56% of all Singapore RTOs) and cross industry transactions (also 56% of all Singapore RTOs). However, looking closely at the sample, only 10 out of 18 transactions on CAT are between cross industry firms.

8 Rank correlation between number of RTO transactions and market performance for Singapore and Thailand is 0.02 and 0.58, respectively.

9

Table 3 Characteristics of RTOs

The sample consists of 59 sample RTOs firms of which only 47 are successful cases and used in further empirical analysis. The statistics reported in Table 3 are for success RTOs only. Deal value is in millions of local currency (LCY). Relative deal size is computed from deal size divided by listed firms’ assets in pre‐MOU year. Premium is pre‐consolidation issue price relative to 3‐month pre‐MOU value weighted average closing price (VWAP) of listed firm. A foreign counter party is defined as a counter party firm outside the listing sample that is incorporated or listed overseas. The table reports both mean and median values. Median values where applicable are shown in parentheses. A firm is categorized as financially distressed if the listed firms’ auditors report says so.

Singapore Thailand Non‐ P‐value Non‐ P‐value Characteristics: All Distress distress diff All Distress distress diff Main board 14 9 5 8 6 2 % Main Board 44% 28% 16% 53% 40% 13% Secondary 18 13 5 7 4 3 % Secondary 56% 41% 16% 47% 27% 20% Same industry 14 7 7 8 3 5 Different industry 18 15 3 7 5 2 % Different industry 56% 47% 9% 47% 33% 13% Deal value (LCY mn) 279 324 151 0.124 2,701 1,185 4,349 0.063 (141) (150) (35.6) 0.086 (1,725) (763) (3,187) 0.032 Relative size 12.17 14.6 5.61 0.213 7.36 7.98 5.48 0.390 (1.96) (2.49) (0.58) 0.079 (6.15) (8.55) (2.97) 0.224 Premium* 42% 35.7% 9.3% 0.517 149% 13.8% ‐15.5% 0.325 (‐6.9%) (11.2%) (‐14.1%) 0.432 (‐7.1%) (‐2.1%) (‐12.7%) 0.245 VWAP (LCY) 0.12 0.08 0.19 0.017 7.86 3.69 9.66 0.396 (0.09) (0.05) (0.15) 0.0008 (1.87) (1.75) (2.00) 0.648 Days from MOU to EGM 260 222 79 65 Day from MOU to completion 321 371 217 88 22 15 7 5 4 1 %Stock swap 69% 47% 22% 33% 27% 7% Stock swap with cash/warrants 10 7 3 10 4 6 %Stock swap with cash/warrants 31% 22% 9% 67% 27% 40% Foreign counter‐ party 18 15 3 None None None % Foreign 56% 47% 9% %EPS growth 3 year pre‐MOU 111% ‐153.1% ‐83.6% 0.542 ‐176% ‐237.3% ‐154.3% 0.463 (‐ (‐79.7%) (‐86.1%) (‐52.4%) 0.075 (‐92%) 232.5%) (‐91.7%) 0.648 %Rev growth 3 year pre‐MOU ‐7.4% ‐11.6% 1.2% 0.433 ‐33.1% ‐46.5% ‐16.2% 0.171 (‐ (‐14.19%) (‐31.5%) (0.03%) 0.086 (‐30.5%) 30.5%) (‐17.4%) 0.196

10

Note: Two outliers are removed from Singapore and Thai sample; namely Kyodo Allied Industries which was offered a tender price four times its VWAP; and E for L which was offered a tender price of THB 25 compared to VWAP of only THB 1.54.

There are some notable commonalities between the characteristics of Singaporean and Thai RTOs. First, the relative deal size computed from deal size divided by listed firms’ assets in pre‐MOU year is larger for distressed samples both Singapore and Thai where the ratios are 14.6 and 7.98 times, respectively. Relative deal size ratios for non‐distressed Singapore and Thai samples are 5.61 and 5.48. We note that deal value in local currency terms of distressed firms are higher than non‐distressed in Singapore transactions, but the reverse is true in Thai samples. In any case, we view that the relative deal size is more important in understanding the bargaining power that outsider firms is likely to have over distressed firms. Second, we note that distressed firm’s premiums are relatively higher than non‐ distressed firms. We define premium as the percentage difference between pre‐consolidation issue price and 3‐month pre‐MOU value weighted average closing price (VWAP) of listed firm. In Singapore RTOs, premiums of distressed firms are 35.7% compared to 9.3% of non‐distressed firms. In Thai RTOs, non‐distressed firms are offered an average of 15% discount their historical VWAP.9 Third, we find that distressed firms have substantially lower average 3‐year pre‐MOU EPS growth rates as well as considerably lower average 3‐year pre‐MOU revenue growth. However, in general neither groups in RTOs samples are doing well in both profitability and revenue generation prospects. For example, in the Singapore distressed and non‐distressed groups, the %EPS growth is ‐153% and ‐84%; whereas the % revenue growth is ‐11.6% and 1.2%. Fourth, we report two date intervals; time in days from MOU date to EGM, and time in days from MOU to completion of capital raising. We focus on days from MOU to EGM which usually marks the consolidated trading of shares of the merged entity. Usually, this takes on average a little over six months or over 200 days for Singapore samples but around two to three months for Thai samples. For some transactions that requires sizable capital raising, the entire process takes up to six months to about one year. Thus, it is clear that RTOs are not exactly a short‐cut in terms of time requirement to list and raise the necessary capital compared to IPOs.

There are also some differences between Singapore and Thai RTO samples in terms of outsider firms of foreign origin and method of payment. From Table 3, more than half of RTO counter parties on the Singapore exchange have foreign origin (incorporated outside Singapore). Some of these foreign firms are located in off‐shore tax havens like British Virgin Islands, however, some are established and even already listed overseas firms.

Based on the characteristics of the data observed, we offer a few analytical points about RTOs. First, unlike what the popular press may suggest, it appears that RTOs are not short‐cuts to getting listed given the length of time to completion. Neither are outsider firms small and of poor quality otherwise the transaction would not have passed the re‐listing requirements of the exchange and the bulk of the transactions would have occurred on secondary boards in particular given Singapore’s listing flexibility on CAT which does not require a minimum market capitalization or profit history as long as there is listing sponsor supporting the firm. Second, some foreign outsider firms prefer to list on the Singapore exchange to increase its international footprint in Asian markets. For example, the Chaswood Group is a large Food & Beverage holding company listed on Bursa Malaysia before signing RTO deal with Asia Silk Group to list on CAT. Another example is India’s leading architectural services firm, RSP Architects

9 It is important not to overly interpret these premiums as being solely in favor for the incumbent listed firms. For example, St. James Holdings for example was offered a tender price approximately the same value as current trading price by outsider firm Perrenial China Trust. However, based on value of property appraisal, the value of the offer is worth 52% discount to appraised NAV of St. James.

11

Planners transaction with Rowsley which is an SGX listed firm engage in renewable energy. Third unlike regular takeovers where the listed target firm have bargaining power enforcing acquirers to submit higher bids (Stulz, 1988; Chiarella and Gatii, 2013), listed firms in reverse takeovers are substantially smaller than the counter party outsiders if we consider the relative deal value of asset size of listed firm is around 11 times and 7 times for Singapore and Thai samples. In any case, distressed firms are offered fairly generous premiums above their VWAPs ensuring that incumbent shareholders can benefit from the share swap.

Finally, we cannot assume that RTO transactions are in general a one‐sided gain in which the outsider firm with stronger prospects and substantially larger size determines the terms limited downside risk for outsider firms otherwise we would probably see transactions predominantly based on share exchange. The use of solely share exchange makes the terms of the exchange contingent on the target and potential synergy value and both incumbent and incoming shareholders both shares gains and losses from the deal (Hansen, 1987). We observe that up to 31% and 67% of the RTO firms in Singapore and Thailand use mixed mode of payment which involves combination of share swap with cash and/or warrants. The use of mixed payment suggests that the outsider firm is unsure of merged value. From our sample, more than half of RTO transactions in Singapore are based on share exchange only. In the Thai sample more firms use combination of share exchange with cash and warrants. Listed firms normally use additional cash raised from new private shareholders to pay for outsider firms’ assets. Thus, we expect the cash component indicates the incoming firm needs to retain immediate value. The warrant issue component helps alleviate the impact of immediate dilution and provide a delayed option to raise capital in the future if things go well.

4. Methods and results

4.1 Announcement reactions and short‐term performance

In this section we measure short‐term reactions to MOU announcements. The short‐term reaction captures the short‐term wealth impact for incumbent shareholders. We also explore how RTO transaction characteristics will have different wealth impact for investors. The standard event study method (Mackinlay, 1997) is used to calculate abnormal return (AR), cumulative abnormal return (CAR) from the market model, and relevant statistics

ARit  Rit  ERit | t  (1)

where ARit , Rit , and ERit | t  are the abnormal, actual, and normal returns respectively. The conditioning information, t is the market return.

12

Figure 1 Plot of cumulative market model abnormal return for RTO announcements (MOU date)

This figure plots the cumulative market model abnormal return for RTO announcements (MOU date). AR R E R |  Define abnormal return as ARit  Rit  ERit | t where it , it , and it t are the abnormal, actual, and normal returns respectively. The conditioning information, t is the market return.

Figure 1 a All sample and by market Figure 1 b Distressed vs non‐distressed

Figure 1 c Low vs high premium Figure 1 d Low vs high relative deal size

Figure 1a plots CARs for all firms, Singapore firms, and Thai firms combined 20 days before and after the MOU date centered at day zero. The plot shows that the market gradually learns about the forthcoming MOU and CAR drifts up to 28.5% by day 20 after MOU date. The plot also reveals the announcement is quite unexpected in Singapore, unlike the Thai market where CARs has already drifted up to almost 19% in days ‐20 to 0. Figures 1b), 1c), and 1d) illustrate CARs of firms separated by distress status, premium level, and relative deal size level. By day 20, we can see that CAR differences is largest with distress status where CAR of firms in distress rises to 43% compared to firms that are non‐distress which sees CAR edging up only to around 10% The CAR of low premiums firms is 21% about half of CAR of high premium firms which is 43%. When firms are separated by relative deal size, the differences in CARs are quite small.

13

Table 4 Cumulative market model abnormal return around event windows

The sample consists of 47 firms on Singapore and Thai exchanges. The table reports cumulative market model abnormal return for RTO announcements (MOU date). Define abnormal return as

ARit  Rit  ERit | t where ARit , Rit , and ERit | t  are the abnormal, actual, and normal returns respectively. The conditioning information, t is the market return. Cumulative abnormal return T k   between days 1 and 2 is derived from CARit   ARit . The standard cumulative abnormal return t T k

CARi 1, 2  is SCARi 1, 2  . T‐test statistics are tests for the null that CAR and SCAR are equal to  i 1, 2 zero. P‐values relating tests of group mean and median differences are in italics.

Event window CAR t‐CAR SCAR t‐SCAR All [‐10, 10] 0.204 3.31*** 1.402 2.46** [‐20, 20] 0.218 2.82*** 1.119 2.63** [‐10, 0] 0.095 2.37** 0.731 2.59** [‐20, 0] 0.085 1.55 0.589 2.44** [0, 10] 0.132 2.22** 1.328 2.15** [0, 20] 0.156 2.74** 1.063 2.44**

Distress [‐10, 10] 0.288 2.99*** 1.796 1.98* Non distress [‐10, 10] 0.084 1.81* 0.796 2.11* Distress‐Non‐distress [‐10, 10] 0.204 1.00 Diff p‐value 0.0673 0.3194 Diff p‐value Wilcoxon 0.0990 0.3370

Low relative deal size [‐10, 10] 0.207 2.40** 1.731 1.68* High relative deal size [‐10, 10] 0.201 2.22** 1.092 1.94* High‐Low ‐0.006 ‐0.64 Diff p‐value 0.5604 0.5918 Diff value Wilcoxon 0.5677 0.5522

Low Premium [‐10, 10] 0.135 1.75* 0.769 2.31** High Premium [‐10, 10] 0.306 3.00*** 2.19 1.97** High‐Low 0.171 1.42 Diff p‐value 0.1798 0.2369 Diff value Wilcoxon 0.2311 0.3391

Table 4 formally tests the significance of CAR and the standardized cumulative abnormal return (SCAR). Since SCAR is derived from CAR divided by the variance of CAR over the event window, extreme observations are given less weight. We first report overall sample averages CAR and SCAR over different windows and provide relevant test statistics (MacKinlay, 1997). The averages reaffirms that the market

14 gradually learns about the news and in general RTO are perceived as good news since both CARs and SCARs are positive and significant at 1% level at windows [‐10, 10] and [‐20, 20]. Over both these intervals, RTO firms’ CAR is slightly above 20%. We also find that the responses are stronger post‐MOU since CAR and SCAR over [0, 10], and [0, 20] intervals are about twice as high (13% and 16%) as those around windows [‐10, 0], and [‐20, 0].

The remainder of Table 4 tests differences in CAR and SCAR when firms are divided by RTO characteristics as identified in Figure 1. Since the interval [‐10, 10] displays strongest reactions, the tests are carried out and reported for this interval only. We find distress firms’ CAR is significantly higher than those non distress. Average CAR and SCAR of distress firms over the event window is 28.8% and 1.79 compared to 8.4% and 0.79 of distressed firms. High RTO premiums firms also exhibit higher CAR and SCAR though the statistical significance is weak. Firms with high or low relative deal size have very similar abnormal return responses confirming Figure 1 d plot.

Overall the short‐term announcement results indicates that the market responds positively to MOU announcements of RTO transactions. Distress firms cumulative returns are substantially higher than non‐distress firms. This could be because they are trading at pre‐MOU VWAP that are very low or because distress firms in general receive higher RTO offer premiums and thus investors look at size of premium as positive signal for the newly merged firm.

4.2 Long term performances of RTO firms

4.2.1 Buy‐and‐hold abnormal returns (BHAR)

Buy‐and‐hold abnormal returns have come the standard of measuring long‐term abnormal performances of corporate announcements (Barber and Lyon, 1997; Khotari and Warner, (2006). BHARs measure the average multi‐period return from a strategy of investing in firms that are affected by an event compared to investment in benchmark portfolios which can be the market or a controlled group of similar non‐event firms. We calculate a one year BHAR for RTO firms in our sample compared to the BHAR of an equal weighted averages buy and hold return of firms in the controlled group. The choice of one‐year is to avoid increasing skewness and upward bias in measuring BHAR that is increasing the holding period. Too long a holding period can cause the result to be confounded by other corporate announcements or market events. We calculate a one‐year BHAR (from day T1 to T2 or 240 trading days) for each RTO firm compared to a benchmark BHAR of non‐RTO firms. Define,

BHAR  T2 1 R  T2 1 ER |  , (2) iT1,T2 ) tT1 it tT1 it it

where the mean BHAR is the equal weighted average of individual firm BHARs and the conditioning information, t is again the market return. The benchmark portfolio is selected by first excluding RTO firms, then we eliminate firms in the top third market capitalization of the Singapore and Thai exchanges. In an independent sort, we allocate firms into decile groups based on annual trading price range. The top decile (most expensive) trading price range are eliminated. Our choice of control group is preferred over a benchmark based on the entire market return which is influenced by performance of large firms. The size of RTO targets are below the median market capitalization of each exchange and they are typically “penny” stocks or are trading the a price range below the median trading price range of the exchange (Pavabutr et. al, 2014). Small trading price range and small caps typically exhibits larger

15 price movements. The screening results in a control firm sample of 501 from the Singapore exchange and 398 from the Thai exchange.

We can assume further that event‐firm abnormal returns are independent as occurrence is quite random and spread out, there is little risk of threat that abnormal returns will be overly‐stated as Mitchell and Stafford (2000) and Brav (2000) point out given a sample size of 47 firms spread out over nine years.

Table 5 Mean BHARs for RTO firms

Panel A of the table reports 12 months BHARs of RTO and controlled portfolio group. BHARs for each group are calculated as the difference between the equal weighted BHARs of event firms and controlled portfolio group. The controlled portfolio group is selected by first separating firms on each stock exchange then eliminated the top third market capitalization firms on both exchanges, Singapore and Thailand. In a separate sort of decile trading price rankings, trading prices in the top decile (most expensive stocks) are deleted. The column two‐sided p‐values report the parametric p‐value and non‐ parametric Wilcoxon p‐values (in parentheses) for differences between RTO sample BHAR and control sample BHAR. The bootstrap p‐value is based on an empirical distribution created by simulating 1,000 pseudo samples. The p‐value is the fraction of mean BHARs from the pseudo samples in larger magnitude than the original RTO sample mean. Panel B reports 12 month buy‐and‐hold returns (BHRs), BHR  T2 1 R 1 and BHAR of firms classified by distressed and non‐distressed status. iT1 ,T2 ) t T1 it Similar p‐values described in Panel A test for differences between distressed and non‐distressed with control sample BHARs.

Panel A

RTO sample Control Sample Two‐sided Bootstrap BHAR BHAR p‐values p‐value Mean 0.006 ‐0.008 0.1992 < 0.0001 Median ‐0.176 ‐0.189 (0.2516) SD 0.816 0.758 Skewness 7.380 9.580 Panel B Two‐sided Bootstrap Distressed BHR BHAR p‐value p‐value Mean 0.0127 ‐0.0601 0.6228 0.366152 Median ‐0.1269 ‐0.242 (0.7245) SD 0.9900 1.025 Skewness 2.157 2.895

Non‐distressed BHR BHAR Mean 0.1878 0.0840 0.0538 < 0.0001 Median ‐0.0596 ‐0.0521 (0.0731) SD 0.6929 0.4849 Skewness 1.643 2.2335

16

Panel A of Table 5 reports 12 months BHARs of RTO and controlled portfolio group. We can see that both the RTO and controlled sample distributions have high positive skewness as usually documented with long‐term abnormal returns. Figure 2a plots the empirical distribution of BHAR of RTO firms illustrating the skewness. The mean BHAR of RTO firms are 0.6% compared to ‐0.8% of controlled firms. The median BHAR of both groups shows overall underperformance relative to the market at large. For the control samples, this is probably due to dominance of smaller firms as they tend to underperform large caps in emerging markets. A two‐tailed test of difference between BHAR of RTO and control sample firm groups using both parametric p‐values and non‐parametric p‐values show that we cannot reject the null that RTO sample firms perform differently from the control sample. The result points that while median RTO firms do not perform differently from controlled non‐RTO firms relative to the entire market, there seems to be a general decline in price impact over the long‐term relative to short‐term accumulated over 20 days post‐MOU.

Figure 2 Empirical and bootstrap distributions of BHARs

Figure 2a plots the empirical distribution of BHARs from sample RTO firms as described in equation 2. Figure 2b plots the bootstrap distribution 1,000 from pseudo samples created by random selection of control firms’ BHAR over the same event period as RTO firms. The vertical lines in Figure 2b are the upper and lower confidence intervals at 5% based on the bootstrap estimates.

17

Given the skewness of the distribution and high return volatility, we also provide statistical inference for the mean BHAR based on a simulated distribution. In a similar framework as in Ikenberry, Lakonishok, and Vermaelen (1995) and Mitchell and Stafford (2000), we assign the evaluation date to a randomly selected control group firms as described in the screening process earlier, we then compute the mean BHAR for this pseudo sample in the same way as the original RTO sample group. This results in one pseudo sample mean. We repeat this step to generate 1,000 BHAR mean and a bootstrap distribution under the null. A bootstrap p‐value in the last column of Table 5 is calculated as a fraction of the mean of the BHAR from the pseudo samples that are larger than the original RTO mean. Figure 2 b provides the distribution plot of the simulated sample mean BHAR. The one‐sided boot‐strap p‐ value indicates that the RTO sample mean of 0.06 is significantly different from the pseudo sample mean.

Panel B of Table 5 reports BHR and BHAR of RTO firms classified by distressed and non‐ distressed status. We find that the average BHR and BHAR of non‐distressed firms are notably larger than distressed firms. BHR of distressed firms are 0.0127 compared to 0.1878 of non‐distressed. This points to the fact that distressed firm stock prices must be declining over 12 month period compared to the short‐run response. In contrast, the BHR of non‐distressed firms remains close to CAR of approximately 20% after the MOU announcement. Once again we report to test if distressed and non‐ distressed group average BHARs are different from controlled group average BHAR as well as the boot‐ strap p‐values. We find significant difference in the non‐distressed group.

4.2.2 What explains differences in BHARs?

The result in 4.2.2 is intriguing. Does the poor average BHAR performance of distressed firms imply that distressed firm RTO are unsuccessful and losing value over time? To answer this question, we run a linear model using the sample of 47 RTO firms. In each OLS model the dependent variable is 12 month BHAR post‐MOU. Explanatory variables in model 1 are those associated with RTO characteristics; namely, relative deal size and premium both of which have shown important in explaining differences in CAR. In model 2, we include on top of RTO characteristics, accounting variables which are one year post‐MOU percentage changes in book value of equity, return on assets (ROA), and net profit margins (NPM). The change in book value of equity controls for possible share offerings post‐ MOU. Since many RTO deals involved capital increases and new share offerings, which will ultimately

18 affect the book value of equity. In fact, on average RTO sample firms sees the number of shares doubled and book value of equity on average increased over 10 folds due to merging with a bigger entity over the course of 12 months post‐MOU. In model 3, we include natural log of turnover (number of shares traded/total number of shares outstanding) and a distress status dummy. We find that three variables which are relative deal size, change in book value of equity, and turnover are most important in explaining differences in BHARs. BHAR is declining in relative deal size and percent change in book value of equity. This is not surprising as large deals relative to listed firm pre‐transaction asset values requires larger capital increase. The correlation between relative deal size ratio and book value of equity increase is 0.86. We do not find premiums having a significant impact on BHAR even though they generate very strong CAR response over the first 20 days of announcements. However, we do find that firms that received high premiums do not experience as much large a change in book value of equity and number of shares. The correlation between premium levels and change in book value is actually negative (‐0.15). The distress dummy does not help explain the differences in BHAR well as this is already reflected in relative deal size and change in equity book value. We believe there is no need to add a country dummy to the model since RTO characters of Singapore and Thai firms are similar as discussed in section 3.

Table 6 Determinants of BHARs

This table reports the coefficients from three OLS models with white corrected t‐statistics. The dependent variable is BHARs of 47 RTO sample firms from Singapore and Thai exchanges. Relative deal size is computed from deal size divided by listed firms’ assets in pre‐MOU year. Premium is pre‐ consolidation issue price relative to 3‐month pre‐MOU value weighted average unadjusted closing price (VWAP) of listed firm. Percentage changes in return on assets (ROA) and net profit margin (NPM) is iss change over one year from MOU. Turnover is computed from average annual turnover (number of shares traded/total number of shares outstanding).

Model 1 Model 2 Model 3 Estimate t‐value Estimate t‐value Estimate t‐value Relative deal size ‐0.0239 ‐2.63** ‐0.0203 ‐2.65*** ‐0.0209 ‐2.94*** Premium ‐0.0387 ‐0.64 ‐0.0352 ‐0.52 0.0159 0.22 %Chg. Book Equity ‐0.0049 ‐6.03*** ‐0.0048 ‐3.47*** ‐0.0037 ‐2.49** %Chg. ROA ‐0.0242 ‐0.27 ‐0.1001 ‐0.91 %Chg. NPM ‐0.0326 ‐0.29 0.0928 0.51 lnTurnover 0.1283 2.06** Distress 0.0706 0.31 AdjRsq 0.1892 0.1429 0.2452 Pr > F 0.0376 0.1203 0.0598 Note: **, and *** denotes statistical significance at 5% and 1% confidence.

4.2.3 Post‐RTO earnings performance

Long term evaluation of corporate announcements using returns is not a straight forward tasks given that we are unable to control other changing information environment as well as the upward bias documented in longer term computations of BHAR. Furthermore, using BHAR to evaluate the success of RTO transactions alone are further complicated the fact that the an RTO event does not really have a definitive completion date because of continuing plan via private offerings, and warrant

19 exercise, and thus the length of holding periods of 6 months, 1 year or more can be deemed arbitrary. Thus, as an alternative to merely examining BHAR, we evaluate long term financial performance of RTO firms 3‐years post‐MOU. As we are still interested in the varying performance between the distressed and non‐distressed group, Table 7 produce a two‐way mean comparison (within groups over different time and between groups over the same time period) between distressed and non‐distressed firms over before MOU and 3 years post‐MOU. The ratios selected measures general liquidity (cash ratio), leverage (debt to assets, and debt to equity ratios), revenue generation and cost control (EPS and NPM), and returns to investments (ROA, and ROE). 10

We find that all firms have lower cash holdings, but distressed firms must have taken more debt as debt to assets has risen even though debt to equity has dropped given large positive change in equity. Despite the book value of equity that has increased, the ROE of distressed firms has increased compared to pre‐MOU periods. We find that distressed firms have significantly inferior NPM and ROA pre‐MOU compared to non‐distressed firms. However, this gap has closed in 3 years after.

10 Only the cash ratio is chose to measure general liquidity since the our RTO samples are spread out over across varying industries 26% in finance and property, 11% in mining, 30% in technology, and the remainder spread out other industries such as consumer, retail, and manufacturing goods.

20

Table 7 Key financial ratios of distressed and non‐distressed firms

This table reports the key financial ratios of RTO sample firms classified by distressed and non‐ distressed status at end of MOU signing calendar year (referred to as before) and 3‐years after MOU signing. ***,**,* denotes significance at 1%, 5%, and 10%, respectively.

Distressed Non‐ distressed Difference t‐stat p‐value Cash ratio Before 0.296 0.204 0.092 1.20 0.238 3 years after 0.113 0.114 0.001 0.00 0.996 Paired mean diff ‐0.169 ‐0.126 t‐stat (p‐value) ‐2.67 (0.018)*** ‐1.95 (0.087)*

Debt to assets Before 0.173 0.237 ‐0.064 ‐0.98 0.332 3 years after 0.251 0.301 ‐0.050 ‐0.96 0.3467 Paired mean diff 0.159 0.107 t‐stat (p‐value) 2.73 (0.016)*** 1.52 (0.166)

Debt to equity Before 1.449 0.698 0.751 1.18 0.2488 3 years after 0.683 0.981 ‐0.298 ‐0.86 0.407 Paired mean diff ‐0.6883 0.5097 t‐stat (p‐value) ‐0.53 (0.609) 1.89 (0.095)*

EPS Before 0.017 0.202 ‐0.186 0.83 0.4193 3 years after 0.067 0.193 ‐0.125 ‐0.81 0.4398 Paired mean diff 0.018 ‐0.251 t‐stat (p‐value) 0.75 (0.473) ‐0.9 (0.399)

Net profit margin Before ‐0.196 ‐0.026 ‐0.170 ‐1.96 0.061 3 years after 0.001 ‐0.007 0.008 0.12 0.9031 Paired difference 0.204 0.054 t‐stat (p‐value) 1.55 (0.181) 0.69 (0.519)

Return on assets Before ‐0.152 0.055 ‐0.207 ‐2.61 0.0129 3 years after ‐0.004 0.061 ‐0.066 ‐1.94 0.0068 Paired difference 0.234 0.005 t‐stat 2.52 (0.026) 0.16 (0.881)

Return on equity Before ‐0.128 0.138 ‐0.266 ‐3.37 0.0022 3 years after 0.014 0.154 ‐0.139 ‐1.86 0.0784 Paired difference 0.207 0.006 t‐stat (p‐value) 1.94 (0.094) 0.09 (0.934)

21

4.3 Liquidity of RTO firms

So far in this paper, we have examined how RTO deals have made an impact on short‐term and long‐term return as well as financial performance. Returns measure the overall impact of investors’ wealth whereas financial performance measures how the transaction has altered the newly merged entity. We finally explore liquidity of RTO firms before and after the transaction marked by the MOU date as before. Liquidity has an important role in reducing the for the firm as Acharya and Pederson (2005) has shown theoretically and empirically in Liu (2006). Liquidity also plays an important part in improving information efficiency as numerous papers in market microstructure argue (see Stoll, 2000; O’Hara, 2003; and Chordia, Roll, and Subrahmanyam, 2008) and thus they are variables that are monitored closely by regulators. Since there are numerous ways to measure liquidity, we select the ubiquitous percentage bid‐ask spreads, and turnover. Bid‐ask spreads are typically strongly associated with price impact or how a change in a unit trading volume can move price. Larger spreads are also associated with asymmetric information. Table 8 report percentage bid‐ask spreads and turnover, 12 months before and 12 months after MOU. The averages are derived from daily data frequency. Both percentage spreads and turnover are significantly reduced. The largest reduction is the Singaporean spread going down to 8.68% from a high of 13.99% pre‐MOU. Turnover has doubled in the Singapore RTO firms increasing from 0.4% to 0.8% whereas turnover rise around 64% for Thai RTO firms. Despite these increases, the post‐MOU liquidity improvement still place RTO stocks in the bottom decile turnover compared to the rest of the Singapore and Thai exchanges (Pavabutr et al. 2014)

Table 8 Liquidity of RTO firms by exchange

This table reports liquidity measures of RTO sample firms. Percentage day‐ end bid‐ask spreads computed from (ask‐bid)/closing price 12 months worth of daily averages and median (in parentheses) before MOU and 12 months after MOU. Turnover is defined as percentage daily average number of shares traded divided by total number of shares outstanding. Daily outlier observations at 1% and 99% are removed. Parametric p‐values and non‐parametric p‐values (in parentheses) are provided.

All Singapore Thailand Measure Liquidity p‐value Liquidity p‐value Liquidity p‐value % Bid‐ask spread 12 months before 10.67% 13.99% 1.39% (5.00%) (8.33%) (1.09%)

12 months after 9.84% 0.0378 8.68% (0.0937) 0.95% < .0001 (2.94%) (< .0001) (5.47%) (< .0001) (0.75%) (< .0001) % Turnover 12 months before 0.82% 0.40% 1.58% (0.08%) (0.03%) (0.38%)

12 months after 1.10% (< .0001) 0.85% < .0001 2.59% 0.0002 (0.10%) (< .0001) (0.03%) (< .0001) (0.65%) (< .0001)

22

Conclusion and policy discussion

Using a sample of 47 firms on Singapore and Thai stock exchanges between 2007‐2015, we seek to understand the motivation to list using reverse takeovers by comparing the RTO to IPO listing process, analyzing the characteristics of RTO listed participants, evaluating the post‐transaction short‐term and long‐term return performance, and examining post‐transaction financial performance. While the popular press often treat these transactions with suspicion as they associate reverse takeovers as a “back‐door” listing process where opportunistic private or outsider firms seek a short‐cut to listing and create speculative hype on small caps, we find no evidence that such parable in our sample RTOs.

Can we regard outsider firms as White Knights? Perhaps so for shareholders of listed firms as we find that RTO transactions provide an exit opportunity for incumbent listed shareholders as they can gain from short‐term returns and improved liquidity of an otherwise forgotten illiquid stock. But do these private investors succeed in rescuing troubled firms and turning the businesses around. Analyzing the post‐transaction financial performance suggests some improvement in distressed firm performance, but not without costs of tremendous dilution from continuous recapitalization. What is not so definitive is why outside firms would want to list this way, if they are already of caliber to list via IPO with their own profile? Our analysis suggests that we should not view outsider firm decisions to conduct RTO as a means to list per se but as a corporate transaction strategy to achieve other ends such as becoming a public company without immediate large dilution in ownership and still have an option to raise more funds later, obtaining a distressed or non‐distressed target public firm at a bargain price especially those with valuable property (see Circular of Perrenial Holdings transactions with St. James holdings in Singapore), and finally, the motivation could be unique to the firms themselves. For example, the Chaswood Group, already listed on Bursa Malaysia decides to reverse takeover Asia Silk Group to list on CAT and explain that the choice of listing on the Singapore exchanges is to increase their international profile in Asian markets. On the Thai exchange, Singha Estate choose to complete a RTO with Rasa property as a short‐cut process to diversify the group’s business from food and beverage into real estate. Going forward, the benefits are costs of RTOs should be studied on a case‐by‐case basis since they are very unique transactions.

Should the regulators loosen requirements on RTOs? Indeed, reverse mergers can be predicted to cause huge percentage‐wise increases in the stock price of a public firm upon the release of news as we might find in any merger news. However, for small cap stocks with very low trading price, even if its stock price rose from virtually zero to only a few cents, the gain could be enormous. Without investor savvy and adequate enforcement, the combination of potentially huge gains and relatively unsophisticated players is a recipe for market abuse. In the end, a totally unobtrusive approach towards RTOs is likely to attract opportunistic Machiavellian behavior.

Fortunately, current rules in place impose the same minimum listing requirement on listings via RTOS as on IPOs. So either way, we expect firms listing directly or indirectly are at par in terms of firm quality. However, if RTOs are to be used as a means to attract potentially strong private firms to help provide rehabilitation, and exit for incumbent public firms then perhaps some concessions should be made on selected listing rules. This observation is probably more applicable in the Thai case as MAI is subject to relatively tighter listing rules with both size and profit requirement unlike CAT listing where neither requirement is present. In addition, it is important to provide investors and media education that RTOs are well regulated transactions and that SEC and Stock Exchange approved deals are not pump‐and‐dump schemes to avoid investor misunderstanding.

23

References

Adjei, F., K. Cyree, and M. Walker, 2008, The determinants and survival of reverse mergers vs. IPOs, Journal of Economics and Finance 32: 176‐194.

Archarya, VV., and Pedersen OH., 2005. Asset pricing with liquidity risk, Journal of Financial Economics, 77(2), 375‐410.

Arellano‐Ostoa, A. and Brusco, S., 2002. Understanding reverse mergers: A first approach, Working paper, Universidad Carlos III de Madrid.

Barber BM., and Lyon JD., 1997. Detecting long‐run abnormal stock returns: The empirical power and specification statistics, Journal of Financial Economics, 43(3), 341‐372.

Berger, A.; Demsetz, R.; and Strahan, P. 1999. The consolidation of the financial services industry: Causes, consequences, and implications for the future. Journal of Banking and Finance, 23 (February): 135–94.

Brav, A., 2000. Inference in long‐horizon event studies: A Bayesian approach with application to initial public offerings, Journal of Finance, 55(5), 1979‐2016.

Carpentier, C., Cumming, D ., and Suret, JM., 2009. The value of capital market regulation: IPO versus reverse mergers, Working Paper, CIRANO.

Chiarella C., and Gatti, S., 2013. How much to pay, and how, for opacity? Negotiating premiums and method of payment in M&A, Working Paper, University of Bocconi.

Chordia, T., Roll, R., and Subrahmanyam, A., 2008. Liquidity and market efficiency, Journal of Financial Economics, 87(2), 249‐268.

Floros, IV., and Sapp, TRA., 2011. Shell games: On the value of shell companies, Journal of , 17, 850‐867.

Floros, IV., and Shastri, K., 2009. A comparison of penny stock initial public offerings and reverse mergers as alternative mechanisms for going public, SSRN.

Gleason, KC., Rosenthal, L., and Wiggins, RA., 2005. Backing into being public: an exploratory analysis of reverse takeovers, Journal of Corporate Finance, 12, 54‐79.

Hansen, R., 1987. A theory of medium of exchange in , Journal of Business, 60, 75‐96.

Hseih, J., Lyandres, E., and Zhdanov, A., 2009. A theory of merger‐driven IPOs, Working paper.

Khotari, S.P., and Warner, J.B., 2006. Econometrics of Event Studies, Handbook of Corporate Finance: Empirical Corporate Finance. Vol A (Handbooks in Finance Series, Elsevier/North‐Holland) Ch. 1.

Ikenberry, D., Lakonishok J., and Vermaelen, T., 1995. Market underreaction to open market share repurchases, Journal of Financial Economics, 39(2), 181‐208.

Liu, W., 2006. A liquidity‐augmented capital asset pricing model, Journal of Financial Econonomics, 82(3), 631‐671.

24

MacKinlay, AG., 1997. Event studies in economics and finance, Journal of Economic Literature, 34, 13‐ 39.

Mitchell, L., and Stafford, E., 2000. Managerial decisions and long‐term stock price performance 73(3), 287‐329.

O’Hara, M., 2003, Liquidity and , Journal of Finance, 58(4), 1335‐1354.

Pakov, AR., 2006. Ghouls and Godsends? A critique of reverse merger policy, Berkeley Business Law Journal, 3(2). 474‐513.

Pavabutr, P., Ghon, R., Sirodom, K., and Tian, G., 2014. Asian pennies are not like US pennies, Development Bank of Japan, Discussion paper series, 1402.

Sjostrom, W., 2008. The truth about reverse mergers, Entrepreneurial Business Law Journal, 2(2) , 744‐ 759.

Stoll, HR., 2000. Friction, Journal of Finance, 55(4), 1479‐1514.

Stulz, RM, 1988. Managerial control of voting rights: Financing policies and the market for corporate control, Journal of Financial Economics, 20(1) 25‐54.

Vermeulen, EPM., 2014. Rules on backdoor listings: A global survey, Indonesia‐OECD Corporate Governance Policy Dialogue.

Wan‐Hussin, WN., 2002. The reverse takeover of Berjaya Textiles and back‐door listing of Jaya Tiasa: A Malaysian case.

Winyuhuttakit, P., 2011. Asset acquisition from unlisted company regarded as backdoor listing, Master of Law Thesis, Chulalongkorn University.

25