The Changing Latitude: Labor-Sponsored Venture Capital Corporations in Canada

The Changing Latitude: Labor-SponsoredVenture Capital Corporations in Canada

Sofia Johan*, Denis Schweizer, and Feng Zhan

ABSTRACT

Manuscript Type: EmpiricalResearch Question/Issue: This paper seeks to understand the role corporate governance and government policy plays in theportfolio choices of the labor-sponsored venture capital corporations (LSVCCs) in Canada. We investigate whether or notthe change in tax policy announced in Ontario (2005) had an impact on the investment behavior of Ontario LSVCCs andwhether the unique corporate governance structure of LSVCCs enables them to focus on their investment mandatesubsequent to this announcement.Research Findings/Insights: Our findings suggest that LSVCCs in Ontario are more likely to include public companies intheir fund portfolios after the announcement of the change in tax policy. We find that after 2005, LSVCCs have increasedtheir number of investments in public companies by 59.13 percent and in turn decreased their number of investments inprivate companies by 13.17 percent. On the other hand, we find no significant changes in investment behavior for LSVCCsin other provinces. In terms of the percentage of total investment in public companies, we find that the LSVCCs in Ontarioare more likely to increase their total investment in public companies by 50 percent and to decrease their investment in theshort term by 46.43 percent. LSVCCs in other provinces, however, are reducing their percentage of investment in publiccompanies by 58.33 percent and increasing their total investment in private entrepreneurial firms by 38.33 percent in thesame period.Theoretical/Academic Implications: With a hand-collected proprietary dataset, we are able to augment existing studies onthe unique structure of LSVCCs in Canada with empirical evidence on the style drift due to the changes in government taxpolicy. We compare and contrast the investment behavior of LSVCCs before and after the tax policy change in Ontario aswell as the investment behavior of LSVCCs in other provinces. We hypothesize that as a result of the elimination of the taxcredits, the removal of certain investment restrictions, and weaker corporate governance, LSVCCs have drifted from theiroriginal mandate to invest in high-risk venture companies to investing in less risky public companies. Such style drift maybe a result of LSVCCs preparing for potential wealth transfer or liquidation by retail investors. More importantly, we findthe unique corporate governance structure of LSVCCs may facilitate this drift from their original purpose of providingventure capital to small and medium-sized entrepreneurial (SME) firms.Practitioner/Policy Implications: We highlight that the style drift of LSVCCs in Ontario may result in such funds behavingmore like other types of mutual funds and the deviation from their original mandate to provide venture capital may not onlyprove detrimental to entrepreneurial investee firms seeking such capital, but also negate any diversification benefits soughtby fund investors. Also, such deviation may not necessarily justify the higher management expense ratio charged byLSVCCs.

Keywords: Corporate Governance, Venture Capital, LSVCC, Tax Policy, Investments

INTRODUCTION

T he economic benefits of venture capital have been wellestablished by numerous academic studies. Research

has established that in addition to bridging the capital gap in

the financing of entrepreneurial firms, venture capital fundmanagers play a significant role in enhancing the value oftheir entrepreneurial investments as they provide consider-able administrative, marketing, and strategic advice toentrepreneurial firms, as well as facilitating a network ofsupport for an entrepreneurial firm with access to accoun-tants, lawyers, and investment bankers (Cumming & Johan,2012a; Gompers & Lerner, 1999; Leleux & Surlemount, 2003;

*Address for correspondence: Sofia Johan , York University – Schulich School ofBusiness, 4700 Keele Street, Toronto, ON, Canada. Tel: 1-647-449-3410; E-mail: [email protected]

145

Corporate Governance: An International Review, 2014, 22(2): 145–161

© 2014 John Wiley & Sons Ltddoi:10.1111/corg.12057

Manigart, Korsgaard, Folger, Sapienza, & Baeyens, 2002;Manigart et al., 2002; Sahlman, 1990; Sapienza, Manigart, &Vermeir, 1996; Wright & Lockett, 2003). To encourageventure capital investment in Canada, Canadian labor-sponsored venture capital corporations (LSVCCs, alsoknown as labor-sponsored investment funds (LSIFs)) werecreated with the objective of providing venture capital tosmall- and medium-sized entrepreneurial (SME) firms tofacilitate economic development and growth. Unlike othermore traditional venture capital funds, the LSVCCs werecreated and structured to raise capital from the large retailinvestor pool and not institutional investors, and to facilitatethis objective, both the federal and provincial governmentsprovide tax credits for individual investors who invest inthese funds. The tax credits are also provided to incentivizeinvestments in such funds as, while they are seen to bepotentially particularly beneficial to the economy in terms ofincreasing innovation and other benefits, they are alsodeemed to be more risky than mutual funds. It is interesting,however, that several studies (Cumming, 2007; Cumming &MacIntosh, 2006, 2007; Sandler, 2004) have found that notonly do LSVCCs provide retail investors with lower returnsbut the performance of LSVCCs are also far behind manyother indices, such as small cap fund indices as well as thelowest risk Treasury bills. These findings cast doubt on thesuccess of the government-sponsored program and questionthe ability of these LSVCCs to provide long-term economicgrowth. More importantly, research suggests that LSVCCshave a negative impact on the overall capital available to theentrepreneurial firms as the tax credits awarded and thefund management mandates or constraints are structuredsuch that the access to cheaper capital and the necessity forsuch capital to be invested within a tight time frame mayhave resulted in a “crowding out” of other types of venturecapital investors (Cumming & Johan, 2012a; Cumming &MacIntosh, 2006). This not only directly undermines theinitial purpose of creating the LSVCCs but may also exacer-bate direct losses in government tax revenue.

As a result, the provincial government of Ontarioannounced on September 30, 2005, that it would graduallyeliminate the tax credit for individual investors investing inLSVCCs. It also lifted the restriction on the investment inpublicly listed companies by LSVCCs. Eliminating the taxcredit would undoubtedly reduce the attractiveness of theventure capital funds to retail investors therefore impactingthe inflow of fund capital into LSVCCs in Ontario. In thispaper, we seek to understand the roles corporate governancestructure and tax policy in Ontario played in the portfoliochoices of the LSVCCs. We investigate whether or not theannouncement of the elimination of the tax credit in Ontariochanged the investment behavior of Ontario LSVCCs andwhether the unique corporate governance structure encour-ages LSVCCs to maintain their original function of provid-ing venture capital to private entrepreneurial firms.

The elimination of tax credits to investors in Ontario willhave an impact on LSVCC fund inflow. Studies have shownthat investors are incentivized to invest in LSVCCs with thepromise of tax credits due to the cheaper actual cost of initialinvestment. More importantly, the same investors are lessconcerned about the long-term after-tax return due to theup-front benefits (Barclay, Pearson, & Weisbach, 1998;

Bergstresser & Poterba, 2002; Cumming & Johan, 2012a;Cumming & MacIntosh, 2003a, 2003b). LSVCCs in Ontariotherefore will become less attractive to retail investors afterthe change in tax policy and there will in turn be addedpressure for higher returns from existing investors as theyget closer to their allowed redemption dates or when theirstatutory lock-up periods expire. LSVCCs have been foundto be underperforming even risk-free 30-day Treasury Bills(Cumming & Johan, 2012a; Cumming & MacIntosh, 2006,2007) and, as a result, fund managers in Ontario areincentivized to alter their investment behavior to counter ormitigate the potential effect of the change in investor senti-ment pursuant to the removal of the tax credit, and thepotential redemptions as the lock-up periods expire. Thefund managers may alter portfolio risk, measured by vola-tility, by drifting away from the less volatile investments inprivate entrepreneurial firms toward more volatile publiccompanies, or the fund manager could continue to provideventure capital to risky firms with high potential upsidereturns (positively skewed returns distribution) as theirinitial investment mandate required. Recall that LCVCCfund managers have distinct skill sets for providing venturecapital and value added to entrepreneurial firms and there-fore they may not be as skilled in investing in publicly listedcompanies as mutual fund managers or hedge fundmanagers. Also, in terms of market risk, volatility is higherfor public equity compared to private investments (seeCumming, Hass, & Schweizer, 2013).

Our findings suggest that pursuant to the announcement,LSVCCs in Ontario are increasingly reallocating their assetsfrom private firms toward public companies. We find thatafter 2005, LSVCCs increased the number of public compa-nies in their portfolios by 59.13 percent and decreased(and/or omissions of reinvestments from exited privatecompanies) the number of private companies by 13.17percent. These results are statistically significant at the 10percent and 1 percent levels, respectively. On the other hand,we find no evidence to suggest the same shift in terms of thefund portfolios for LSVCCs in other provinces within thesame period. In terms of percentage of total investment inpublic companies, we find that while the LSVCCs in Ontarioare more likely to increase their investments in public com-panies by 50 percent, the LSVCCs in other provinces havedecreased their holdings in public companies by 58.33percent since 2005, when compared with the overall average.Our results also indicate that LSVCCs in provinces otherthan Ontario are likely to increase their total investmentin private entrepreneurial firms by 38.33 percent whileLSVCCs in Ontario are more likely to decrease their totalinvestment in the short term by 46.43 percent comparedwith the overall level of average investment in short term.

Venture capital funds are known to style drift, for examplefrom the riskier early stage to the latter stages of investmentsuch as mezzanine and buy-out stages, and they are alsoknown to invest in public versus privately held firms(Chaplinsky & Haushalter, 2010; Cumming, Fleming, &Schwienbacher, 2008; Cumming & Johan, 2012a). Ourresearch augments such academic work as our results indi-cate that after the changes in tax policy in 2005, there is atrend for LSVCCs in Ontario to drift away from investing inprivate entrepreneurial firms and increasingly investing in

146 CORPORATE GOVERNANCE

Volume 22 Number 2 March 2014 © 2014 John Wiley & Sons Ltd

public companies, both in terms of the number of publiccompanies in the fund portfolio and the percentage of totalinvestment in the fund portfolio. LSVCCs in Ontario may bestyle drifting away from providing venture capital andmoving toward the more liquid investment behavior ofother types of mutual funds more commonly utilized byretail investors as investment vehicles. Considering thatLSVCCs were created with the intention of facilitating theprovision of venture capital to entrepreneurial firms, ourresults suggest that LSVCCs may not be fulfilling theirintended function.

LSVCCs were also created to enable retail investment inthis alternative investment vehicle that is not ordinarilyavailed to investors other than institutional and high networth investors. This alternative investment also enabledretail investors to obtain portfolio diversification benefits,and as such incentives in the form of tax credits were pro-vided by both the federal and provincial governments tomake such investments more palatable. Our results alsosuggest that style drifting by Ontario LSVCCs managersmay be misleading retail investors as they would have ini-tially acquiesced to the higher management expense ratios(as compared to mutual funds) because they had invested ina venture capital fund, not a fund similar in strategy to othermutual funds. In addition to not fulfilling their function,they may also not be executing their investment mandates asagreed upon with their investors. Furthermore, we find thatthe return on assets (ROA) of LSVCCs in Ontario havedecreased statistically significantly after the changes in taxpolicy were announced, while the volatility simultaneouslyincreased. Both results fit into the view that the style driftmay by increasing overall portfolio risk of both LSVCC andinvestor, and not reducing it.

More importantly, our study provides new evidence of theeffect of weak corporate governance on LSVCCs. Retailinvestors are unlike other institutional investors in that theirdisparate make-up does not allow for efficient oversight offund managers. That, in addition to the disincentives foroversight or demands for excellent returns as a result of taxcredits, may enable or even encourage fund managers tostyle drift, take on additional portfolio risk and move awayfrom their original function of providing venture capital.

Our paper contributes to extant literature in the followingmanner. First, we are the first to examine the effects of achange in government policy on a retail-based government-sponsored venture capital program. Our study not only ana-lyzes the affects of tax policy changes on LSVCCs in onejurisdiction, Ontario, but also compares the LSVCCs inOntario to others across multiple jurisdictions or rather tothose situated in and governed by other provincial govern-ments. Second, we provide empirical evidence that theannouncement of the removal of tax credits in Ontario hashad an impact on the investment behavior of LSVCCs inOntario. Pursuant to the announcement, LSVCCs in Ontarioare increasingly shifting away from venture capital invest-ments in private entrepreneurial firms and investing more inpublicly listed companies. Third, we do not find similardrifts among LSVCCs across other jurisdictions or provincesand therefore we believe our results have several policyimplications. The central purpose of creating LSVCCs was tobridge the capital gap faced by entrepreneurial firms and

facilitate the provision of both financial capital and also themore significant value added provided by venture capitalfund managers. In addition, the creation of LSVCCs enabledretail investors to invest in an alternative investment assetclass (venture capital) that would ordinarily only be availedto institutional investors and high net worth individuals,and as a result potentially diversify from their portfolio ofmutual funds, stocks, and bonds. If LSVCCs are style drift-ing away from venture capital and increasingly shifting theirinvestments toward public companies without providingsuperior returns, as supported by our findings, this raisesthe question of whether or not LSVCCs are fulfilling theireconomic function, executing their mandates as agreedupon with their investors, and generally doing what they arepaid to do.

The rest of the paper is organized in the following manner.The next section provides a brief description of characteris-tics of LSVCCs in Canada and provides an overview of thegeneral tax credit provided by both federal and provincialgovernments to individual retail investors. We then discusssome theoretical arguments that affect fund manager invest-ment behavior due to the change of government tax policyand introduce our three hypotheses. The data and summarystatistics are then introduced, followed by sections present-ing the univariate comparison and the multivariate analysesof the impact on the announcement of the change in the taxpolicy in Ontario. We then provide some robustness checks,and concluding remarks follow in the last section.

LSVCCS IN CANADA AND OVERALL TAXCREDITS TO LSVCC INVESTORS

The development and characteristics of LSVCCs, theprimary government support mechanism for the provisionof venture capital in Canada since the 1980s, have been thor-oughly researched in several studies (Cumming & Johan,2012a, 2012b; Cumming & MacIntosh, 2003c, 2005, 2006).Here, we briefly summarize the unique characteristics ofLSVCCs that were created and structured with the principalpurpose of increasing the pool of capital to finance smallentrepreneurial firms and to promote local economic devel-opment. The first LSVCC was formed in the province ofQuebec in 1983. Other provinces in Canada were quick tofollow with their own legislation to form their own LSVCCs.

While the LSVCCs across provinces may not be identicalin both legal and operational structure, there are somecommon characteristics among all LSVCCs across Canada.First, unlike other venture capital funds that are usuallyformed as limited liability partnerships, LSVCCs are formedas corporations. Second, similar to mutual funds and unlikeother forms of alternative investments, investors in LSVCCsare individual retail investors, and not sophisticated inves-tors such as institutional investors or high net worth indi-viduals. Unlike mutual funds where there is a minimumamount required to invest, there is no such minimumrequired to invest in LSVCCs. Third, individuals investing inLSVCCs receive tax benefits from both the federal and pro-vincial governments. For example, the federal governmentprovides a tax credit of 15 percent. The provincial govern-ments also provide a tax credit although amounts may

THE CHANGING LATITUDE: LSVCCS IN CANADA 147

Volume 22 Number 2 March 2014© 2014 John Wiley & Sons Ltd

differ. In Ontario, for example, the provincial governmentalso provides an additional 15 percent provincial tax creditin addition to the federal tax credit. Fourth, all LSVCCsrequire at least one member of the board of directors to befrom a labor union, hence the labor sponsorship. In addition,the provincial and/or federal legislations often limit boththe types of companies and the amount of investment thatthe fund can invest in each company. Last, LSVCCs oftenhave geographic restrictions on the location of both investeefirms and investors. Due to its structure, Cumming andMacIntosh (2006) and Cumming and Johan (2010, 2012a)argue that the resulting agency problems and costs have leadto poor investment returns.

With many commonalities among different LSVCCs,there are also many differences. First, many provincialLSVCCs are only allowed to accept investments from thelocal, provincial residents except for LSVCCs formed inQuebec or federally incorporated LSVCCs. Second,minimum holding periods or lock-up periods vary in differ-ent provinces. For example, LSVCC investors may onlyexecute redemptions 8 years after initial investment. Mani-toba legislation, however, requires a minimum lock-up of 7years and in Quebec, investors will not be able to redeemtheir shares until their retirement.

To support the development of LSVCCs in Canada, bothfederal and provincial governments provide generousincome tax credits for individual investors. In Ontario, forexample, prior to 2007, individual investors were able toclaim a 15 percent tax credit on a maximum investment up toCan$5,000.After 2007, the maximum investment amount wasincreased to Can$7,500. This means that individuals are ableto claim up to Can$750 in tax credits before 2007 and up to$1,125 in tax credits from 2007 and beyond.1 To furthersupport investment in innovation, the Ontario governmentalso provides an additional 5 percent tax credit if investmentis made in research-oriented investment funds. The federalgovernment provided a similar benefit in income tax benefits.In other words, individual investors could enjoy up to animmediate 30 percent tax benefit (returns) after their invest-ment, even before any investment profits are accounted for.In effect, depending on the individual’s tax bracket, the realcost of a Can$5,000 investment in LSVCCs through a Regis-tered Retirement Savings Plan (RRSP) (a type of governmentregistered savings plan which allows account holders to savefor their retirement on a tax-deferred basis) may only rangefrom Can$1,180 to Can$2,390 (Cumming & MacIntosh, 2006).The tax credits not only made the high-risk investment in theventure capital funds more attractive, but arguably it negatedthe need for individual investors to be overly attentive overtheir investment (as RRSP investments are for retirementpurposes, hence long term in horizon).

In September 2005, the Ontario government announcedthat they intended to remove the tax credit for investmentsin LSVCCs at the end of the 2010 tax year. In May 2008, theOntario government delayed such process and planned toeliminate all the tax credits by the end of the 2011 tax year.2The plan was to gradually cut the tax credit by 5 percent eachyear, starting in tax year 2010, and to totally eliminate the taxcredit by the tax year 2012. Also, from 2005, the LSVCC wasonly required to invest 60 percent of capital raised in privateentrepreneurial firms or SME companies instead of the 70

percent as originally required. In the next section we willdiscuss why we believe such reform will result in changes toLSVCC behavior within Ontario.

LSVVC GOVERNANCE STRUCTURE ANDFUND MANAGERS’ INVESTMENT

BEHAVIOR

We believe that there are at least two theoretical reasons toexplain how eliminating the tax credits could affect the fundmanagers’ investment behavior. First, individual investorsare incentivized to invest with a promise of tax credits due tothe cheaper actual cost of initial investment and, moreimportantly, are less concerned about the long-term after-taxreturn due to the up-front financial benefits (Cumming &Johan, 2012a; Cumming & MacIntosh, 2003a, 2003b). WhileBarclay et al. (1998) argues that exiting fund investors preferto defer the taxable capital gains, Bergstresser and Poterba(2002) find that after-tax returns affect investors’ investmentdecisions and explain better net mutual fund inflow. Beforethe tax reform, LSVCCs in Ontario benefitted from the sub-stantial tax credits (up to 30 percent) so much so that anegative portfolio (pre-tax) return could potentially turninto a substantial positive return after tax savings for itsinvestors. Cumming and MacIntosh (2003a, 2003b) showthat investors lower their required rate return from LSVCCinvestment when compared with both mutual funds orprivate equity fund investment. It follows, therefore, thateliminating the tax credit will not only make LSVCCs lessattractive for future investors, but existing investors willseek higher required rates of return for their investment. Assuch, the fund managers may change portfolio risk, mea-sured by volatility, by drifting away from the less volatileinvestments in private entrepreneurial firms toward morevolatile public companies, or the fund manager could con-tinue to provide venture capital to risky firms with hugepotential upside returns (positively skewed returns distri-bution) as their initial investment mandate required.

Hypothesis 1: After the announcement of the elimination of thetax credit, LSVCCs in Ontario will style drift, while LSVCCsin other provinces will not.

Second, the special tax treatment given to LSVCCs hasresulted in a competitive advantage over other forms ofinvestment funds availed to retail investors such as mutualfunds. With the removal of the tax benefits, the pre- andafter-tax returns expectations will converge. Given the rela-tively poor average performance of LSVCCs in Ontario, thiswill most likely translate into an underperformance, withexpected withdrawals from investors as a result. Anticipat-ing this scenario, LSVCC managers in Ontario have an incen-tive to change their investment behavior and increaseportfolio risk in order to improve expected performance (notnecessarily risk-adjusted performance) and increase liquid-ity. Similar behavior by mutual fund managers is referred toas tournament behavior (see, for instance, Brown, Harlow, &Starks, 1996; Chevalier & Ellison, 1997).

Similar tournament behavior within LSVCCs in Ontariomay be demonstrated as a result of various factors. Rather



than using detailed, highly frequent and timely data, indi-vidual investors tend to pay more attention to annual per-formance rankings. As there are lock-ups in place, and mostLSVCC investments are made within registered retirementportfolios to maximize tax benefits, individual investors maynot necessarily pay attention to such rankings on an annualbasis. Empirical evidence has also shown a convex relation-ship between fund flows and past performance for mutualfunds (Del Guercio & Tkac, 2002; Sirri & Tufano, 1998).“Underperformers” tend to be less affected by outflows than“outperformers” are affected by fund inflows. As a result oftax benefits, long-term investment horizons, and lock-ups,such relationship between fund flows and performance maybe similar for LSVCCs. LSVCCs are similar to mutual fundsin that individual investors are allowed to invest, albeit withmaximum investment restrictions (typically up to $5000),and this results in a highly dispersed investor structurewhich apparently reduces the incentive to closely monitorfund managers due to coordination costs and free-riderproblems (see Cumming & MacIntosh, 2006).

The inclusion of the restrictive lock-ups of up to eightyears also clearly reduces the incentive to monitor fundmanagers as investors are unable to redeem their invest-ments in the event of unsatisfactory performance. Also, theLSVCC structure is designed as open-ended instead of themore standard, limited life-time of about ten years typical ofother venture capital funds. The ten-year time limit essen-tially puts in place a schedule whereby fund managers mustmeet agreed upon performance measures related to invest-ments, monitoring and mentoring of investees, and exits bythe end of the life of the fund. Fund managers that are unableto do so will face extreme difficulties raising capital forfollow-up funds. Open-ended LSVCCs have no equivalenttime frame, other than the lock-ups, to provide suchdiscipline.

Finally, each LSVCC fund is required to have a labor unionsponsor or director. The labor sponsorship was imple-mented to ensure that the LSVCCs remain faithful to theinitiative to create innovate provincial companies that wouldfoster employment, not to necessarily provide value-addedassistance to fund managers. Unfortunately such provincial-centric sponsorship by unions with no ownership interest inthe funds themselves may have only served to hinder theLSVCCs in their investment activities, and more specificallyprovided no added corporate governance other than toensure the maintenance of statutory restrictions (Cumming& MacIntosh, 2003c). Less frequent monitoring by retailinvestors, labor union board members, and disproportionatefund flows can only impede effective fund manager gover-nance and encourage a drift in investment behavior. Recallthat LSVCC fund managers are experts in investing inventure capital, and any drift away from this form of capitalmay not necessarily benefit fund investors. Therefore, ifinstead of continuing to provide venture capital to riskyfirms with huge potential upside returns (positively skewedreturns distribution) as their initial investment mandaterequired, thus maximizing risk-adjusted returns, the LSVCCmanagers in Ontario could start drifting away from the lessvolatile investments in private entrepreneurial firms towardmore volatile public companies taking on added risk(increase in return distribution’s variance) in the hope of

improving their relative performance compared to otherLSVCCs or mutual funds as well as their competitors. Suchmove from venture capital to public equities could also bemade to ensure sufficient liquidity in the event of potentialredemptions. Regardless of motive, such tournament behav-ior may not necessarily benefit existing investors althoughthe rather restricted corporate governance as explainedabove may actually facilitate it. This leads to our secondhypothesis:

Hypothesis 2: After the announcement, LSVCCs in Ontariowill increase their investments in public companies andincrease their risk levels (measured by volatility), whileLSVCCs in other provinces will not.

In addition, it has to be noted that only Ontario has carriedout such tax reform with the removal of the tax credits.Across other Canadian provinces, LSVCCs were createdwith a similar common objective of providing venturecapital to SME firms to facilitate economic development andgrowth, and tax credits to individual investors were alsoprovided. Unlike the government of Ontario, other prov-inces continue with their support of their LSVCCs. Thisenables us to carry out a multi-jurisdictional comparison ofbehavior among fund managers in Canada. Our thirdhypothesis is developed based on this discussion:

Hypothesis 3: LSVCCs in Ontario are expected to drift awayfrom their original mandates, while LSVCCs in other provincesare expected to hold or increase their venture capital investmentlevels in private firms.

DATA DESCRIPTIONS ANDSUMMARY STATISTICS

The data used in this paper comprises 41 LSVCCs acrossCanada. Like many other types of investment funds,LSVCCs are required to publish audited annual reports totheir shareholders. We reviewed all available LSVCC annualreports and hand collected the data based on informationobtained from these reports. The data comprises informationfrom 1997 to 2011 and the main dependent variables are thenumber of public companies in the fund portfolios at the endof the fiscal year versus the number of private firms in thefund portfolios at the end of the fiscal year. We also measurethe percentage of investments, as the percent of overallinvestment, in public companies, at market value, percent-age of investment in private firms at fair market value, andpercentage of short-term investment or marketable securi-ties of fixed-term investment with a term maturity of lessthan one year at market value.

Cumming (2006) finds both fund and investee character-istics have an influence on the portfolio size of venturecapital funds. We obtained several proxies for LSVCC fundcharacteristics from their annual reports, including FundAge which is the number of years since the LSVCC incorpo-rated up to 2011, the Management Expense Ratio which isthe total fees paid to fund managers and other parties as apercentage of the total assets of the fund, the Log (TotalAssets) which is the log of the fund total assets in US$



millions in the lagged period, and the Return on Assets(ROA) which is the net income over total assets at the end ofthe year. We also obtained the type and stage of investeefirms from various sources. A dummy variable equal to 1 isused to indicate Early Stage firms, and the same dummyvariable is used to indicate High-Technology firms. As addi-tional control variables, Cumming and MacIntosh (2006)find that both bank interest rates and overall stock marketperformance have significant impact on the supply as well asthe demand of the LSVCC funds. We obtained the bankinterest rate in the lagged period from the Bank of Canada.Similar to the proxy used in Cumming and MacIntosh(2006), we gather the SP/TSX return in the lagged periodfrom Bloomberg and calculate the index annual returns. Wealso collect log of provincial GDP growth data from StatisticsCanada.

Figures 1 and 2 illustrate the LSVCC’s investment behav-ior over time. Prior to 2001, LSVCCs hardly invested in anypublic companies. After 2001, we see a gradual increase inthe number investments in public companies in the portfo-lios. We especially observe a clear upward trend in terms ofthe number of public companies in the portfolios forLSVCCs in Ontario. There is also an increase in number of

investments in public companies in the portfolios ofLSVCCs in other provinces, but we observe a sharp decreaseafter 2010. From Figure 2, we find that LSVCCs in otherprovinces invest in more private companies than LSVCCs inOntario.

Table 1 provides descriptive statistics for the full sample ofthe data. The summary statistics show that LSVCC fundsinvest heavily in private firms both in terms of number andin the percentage of total amount invested. On average,LSVCCs invest in approximately 21 private companies andweight over 60 percent of total investment in the privatefirms. This is not surprising as the primary objective ofLSVCCs is to provide venture capital to private entrepre-neurial firms. In addition, the data shows that LSVCCs arerewarded higher management fees. We find that the averagemanagement expense ratio is 3 percent (more than doublethat of mutual funds). This number is very similar to the 4percent found in the study by Cumming and MacIntosh(2006). Extant research (Cumming & MacIntosh, 2006;MacIntosh, 1997; Osborne & Sandler, 1998) has found thatLSVCCs provide lower returns relative to other types ofinvestment vehicles. In our data, we find that on average, thereturn for LSVCCs in Canada is −4 percent per annum.

UNIVARIATE TESTS

Table 2 provides a comparison of means and medians testsof the number of public companies; number of private firmsin the portfolios; and Return on Assets (ROA) before andafter the announcement of the change in tax policy in 2005,respectively.

Panel A of Table 2 reports the difference in means andmedians of the number of public companies in the fundportfolio in general; in the Ontario LSVCC funds; and in thefunds of all other provinces, respectively. The mean(median) value of the number of public companies in thefund portfolios are 1.04 (0) before the announcement of taxpolicy changes in Ontario in 2005 (before 2005, hereafter)and 6.80 (1) after the announcement of tax policy changes inOntario (after 2005, hereafter). Our comparisons show thechanges in the number of public companies in the fundportfolios in terms of both mean and median value are sta-tistically significant at the 1 percent level. The picture is evenclearer after we separate the funds by their location: LSVCCfunds that are in Ontario and LSVCC funds that are in allother provinces. For funds in Ontario, the mean (median)number of public companies in the portfolios are 1.07 (0)before 2005 and 8.20 (2) after 2005. There is a drift fromprivate companies and a clear increase of investment inpublic companies for the Ontario LSVCCs, which supportsHypothesis 1 and Hypothesis 2. Our statistical tests supportthis view. The statistical tests for difference in mean andmedian show that this is significant at 1 percent for both themean and median numbers. For funds in other provinces,the mean (median) number of public companies in the port-folios are 1 (0) before 2005 and 3.44 (0) after 2005. The sta-tistical tests show that although there is an increase ofinvestment in the public companies for LSVCCs in otherprovinces, it is only significant in the mean but not signifi-cant in the median.

FIGURE 1Average Number of Public Companies in the Fund

Portfolio between 1997 and 2011

0

2

4

6

8

10

12

14

16

18

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Overall Funds in Ontario Funds in other provinces

FIGURE 2Average Number of Private Firms in the Fund Portfolio

between 1997 and 2011

0

20

40

60

80

100

120

140

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Overall Funds in Ontario Funds in other provinces



Panel B of Table 2 shows the average (median) value is20.48 (13) before 2005 and 20.81 (16) after 2005 for thenumber of private companies in the fund portfolios. Sepa-rating funds by their location, the mean (median) value is15.49 (12.5) before 2005 for funds in Ontario and 30.75 (18)for funds in other provinces before 2005. After 2005, themean (median) value is 18.19 (16) for funds in Ontario and27.05 (15) for funds in the other provinces. From the statis-tics, we observe a slight increase in the number of privatecompanies in the fund portfolios in Ontario and a slightdecrease in the number of private companies in the fundportfolios in other provinces. However, none of our statisti-cal tests come up to be statistically significant.

Panel C of Table 2 shows the average (median) value ofReturn on Assets (ROA) for the full sample of data. Onaverage (median), the LSVCCs provide −2 percent (−1percent) before 2005 and −5 percent (−2 percent) after 2005.Specifically, funds in Ontario on average (median) have anROA of −1 percent (−1 percent) before 2005 and −6 percent(−2 percent) after 2005. The difference is at 10 percent sig-nificance for the average test. For funds in other provinces,the average (median) ROA is −2 percent (−1 percent) before2005 and −3 percent (−1 percent) after 2005. However, wecannot find any statistically significant difference betweenpre-2005 and post-2005 for funds in other provinces. Never-theless, we find that the risks (in terms of standard devia-tion) in returns are much greater for funds in Ontario after2005, the same as our prediction in Hypothesis 2. On thecontrary, we find that the risks in return decrease for fundsin other provinces from before 2005 to the after 2005 period.

In line with Hypotheses 1 and 2, the comparison resultsshow that LSVCCs in Ontario have drifted from investing in

private companies and have invested in more public compa-nies after the announcement of changes in tax policies andhave undertaken more risk after the announcement as well.This type of univariate comparison is not fully informative,as we have not yet controlled for other factors to be equal,particularly in regard to both economic and market condi-tions across different provinces. Our empirical tests in thenext section include variables that have been found to haveinfluences on fund investment and return.

Table 3 presents a correlation matrix for the main variablesused in the multivariate tests. From the correlation matrix,we find a positive relationship between the investment inpublic companies and the change in tax policy in Ontario. Inaddition, the correlation matrix also shows that there is astatistically significant correlation structure among severalvariables. In this case, we present alternative specificationswith and without collinear variables in the regression.

MULTIVARIATE ANALYSES

To understand the impact of the changes in the tax policywith respect to investment strategy, we conduct empiricaltests to show that there is a change in the components of theportfolios after 2005 (see Table 4). We test the same eightmodels with and without controlling for investee character-istics, fund characteristics, and macroeconomic variables forboth panels in Table 4. The dependent variable is the numberof public companies in the portfolios for Panel A and thedependent variable is the number of private firms in theportfolios for Panel B. Models 1 and 2 run a simpledifference-in-difference (DID) regression on the tax policy

TABLE 1Descriptive Statistics

Variable name Mean Median Standarddeviation

Minimum Maximum No. ofobservations

No. of public companies in the portfolios 4.16 .00 11.54 .00 83.00 244No. of private firms in the portfolios 20.66 15.00 22.18 .00 132.00 242% Investment in public companies .12 .00 .25 .00 1.00 244% Investment in private firms .60 .61 .24 .00 1.00 244% Investment in short term .28 .28 .23 .00 1.00 244Treat .68 1.00 .47 .00 1.00 244After 2005 .54 1.00 .50 .00 1.00 244Treat * after 2005 .38 .00 .49 .00 1.00 244Early stage .25 .00 .44 .00 1.00 244Technology .41 .00 .49 .00 1.00 244Fund age 6.01 6.00 4.02 .00 18.00 244Management expense ratio .03 .02 .14 .00 2.24 241Log (Total Assets) 17.56 17.65 1.45 10.52 20.44 215Return on Assets (ROA) −.04 −.01 .17 −1.15 .35 244Bank rate 3.27 3.19 1.26 .65 5.77 244SP/TSX return .06 .12 .21 −.35 .31 244Log (Provincial GDP Growth) .02 .02 .02 −.03 .08 238

This table present statistics for the full sample of fund-year observations in the data. The data range from 1997 to 2011.



change in 2005 with different cluster controls. Models 3 and4 show the results with control on investee characteristicsand fund characteristics, respectively. Model 5 shows theresults with macroeconomic variables and Model 6 presentsthe results with all control variables. For the last two models,

Models 7 and 8, we repeat Model 6 and test the results withfunds in Ontario and funds in other provinces, respectively.

The results in Panel A of Table 4 consistently show thatafter the change in tax policy in 2005, LSVCC funds are morelikely to include public companies in their portfolios. The

TABLE 2Comparison Tests

Panel A: Number of public companies in the portfolios

All funds Ontario funds Other province funds

Before 2005 After 2005 Before 2005 After 2005 Before 2005 After 2005

Group 0 1 0 1 0 1No. of observations 112 132 74 93 38 39Mean 1.04 6.80 1.07 8.20 1.00 3.44Median 0 1 0 2 0 0Standard deviation 3.80 14.81 4.44 16.87 2.08 7.11Difference in mean (0–1) −4.298*** −3.913*** −2.050**Difference in median (0–1) p = .000*** p = .000*** p = .112

Panel B: Number of private firms in the portfolios



Group 0 1 0 1 0 1No. of observations 110 132 74 93 36 39Mean 20.48 20.81 15.49 18.19 30.75 27.05Median 13 16 12.5 16 18 15Standard deviation 22.49 22.00 13.45 16.32 32.21 31.09Difference in mean (0–1) −0.114 −1.175 0.505Difference in median (0–1) p = .411 p = .239 p = .213

Panel C: Return on Assets (ROA)



Group 0 1 0 1 0 1No. of observations 112 132 74 93 38 39Mean −.02 −.05 −.01 −.06 −.02 −.03Median −.01 −.02 −.01 −.02 −.01 −.01Standard deviation .12 .20 .11 .23 .13 .10Difference in mean (0–1) 1.796* 1.915* 0.144Difference in median (0–1) p = .304 p = .188 p = .910

This table presents the comparison of mean and median tests for the number of public companies in the portfolios (Panel A), number ofprivate companies in the portfolios (Panel B) and Return on Assets (Panel C).*Significant at the 10% level**Significant at the 5% level***Significant at the 1% level



TA

BL

E3

Cor

rela

tion

Mat

rix

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(1)

No.

ofpu

blic

com

pani

esin

the

Port

folio

s1

(2)

No.

ofpr

ivat

efir

ms

inth

ePo

rtfo

lios

−.10

51

(3)

%In

vest

men

tin

publ

icco

mpa

nies

.808

*−.

231*

1

(4)

%In

vest

men

tin

priv

ate

firm

s−.

477*

.223

*−.

545*

1

(5)

%In

vest

men

tin

shor

tte

rm−.

388*

.023

1−.

525*

−.42

8*1

(6)

Trea

t.1

11−.

265*

.014

2.0

68−.

0843

1(7

)A

fter

2005

.241

*−.

0342

.243

*.1

83*

−.44

8*.0

751

1(8

)Tr

eat

*af

ter

2005

.254

*−.

154*

.190

*.1

34−.

341*

.562

*.7

33*

1(9

)E

arly

stag

e−.

140*

−.20

8*−.

0779

.001

73.0

822

.042

2−.

117

−.09

721

(10)

Tech

nolo

gy−.

173*

−.02

75−.

112

.153

*−.

035

−.13

3−.

0369

−.15

0*.0

434

1(1

1)Fu

ndag

e.1

26.2

12*

.125

.211

*−.

349*

.021

9.4

57*

.325

*−.

246*

.007

31

(12)

Man

agem

ent

expe

nse

rati

o−.

0061

1−.

0569

−.01

39−.

0688

.084

8.0

288

−.08

64−.

0744

−.05

56−.

0567

−.09

71

(13)

Log

(Tot

alA

sset

s)−.

132

.613

*−.

186*

.251

*−.

0545

−.04

07.0

094

−.00

03−.

0765

.036

6.2

68*

−.36

2*1

(14)

Ret

urn

onA

sset

s(R

OA

).0

39.0

1.0

672

−.11

4.0

432

−.06

02−.

0947

−.13

1−.

112

−.18

6*−.

0636

.010

2−.

0351

1(1

5)Ba

nkra

te−.

183*

.129

−.19

6*−.

0368

.248

*−.

032

−.11

4−.

0734

.069

8.0

639

−.24

8*−.

041

.203

*−.

223*

1(1

6)SP

/T

SXre

turn

.022

9−.

0061

.040

6−.

0598

.017

−.06

42−.

0459

−.08

18−.

0492

−.09

41−.

117

−.05

15−.

0173

.238

*−.

166*

1(1

7)L

og(P

rovi

ncia

lGD

PG

row

th)

−.19

3*.0

236

−.13

9*−.

0932

.244

*−.

157*

−.38

9*−.

393*

.029

9.0

304

−.28

6*.0

366

.054

.121

.340

*.2

76*

1

This

tabl

epr

esen

tsPe

arso

nco

rrel

atio

nco

effic

ient

sfo

rth

efu

llsa

mpl

eof

fund

-yea

rin

the

data

.*S

igni

fican

tat

the

5%le

vel.



TA

BL

E4

Reg

ress

ion

Res

ult

s

Pane

lA:N

umbe

rof

publ

icco

mpa

nies

inth

epo

rtfo

lio

Mod

el1

Mod

el2

Mod

el3

Mod

el4

Mod

el5

Mod

el6

Mod

el7

Mod

el8

DID

DID

Inve

stee

char

acte

rist

ics

Fund

char

acte

rist

ics

Mac

roec

onom

icva

riab

les

All

vari

able

sO

ntar

iofu

ndon

lyO

ther

prov

inci

alfu

nds

Con

stan

t1.

00**

*1.

03*

2.95

25.1

39**

*6.

14**

*28

.13*

*33

.68*

−7.2

6[3

.33]

[1.7

1][1

.65]

[3.1

4][4

.17]

[2.3

5][2

.01]

[−.4

4]Tr

eat

.07

.04

.41

.050

.01

.38

[.30]

[.11]

[.30]

[.12]

[.02]

[.83]

Aft

er20

052.

44**

2.91

**3.

34*

2.07

2.00

**2.

60**

4.14

*−1

.23

[2.9

4][2

.08]

[1.9

7][1

.03]

[2.4

3][2

.24]

[1.8

2][−

.61]

Tre

at*

Aft

er20

054.

70**

*4.

22**

*3.

00**

4.25

**4.

08**

*2.

46*

[3.8

2][1

4.48

][2

.31]

[2.3

6][2

.98]

[1.8

8]In

vest

eech

arac

teri

stic

sE

arly

stag

e−2

.75*

*−3

.34*

**−4

.83*

*2.

35**

[−2.

25]

[−3.

79]

[−3.

71]

[2.8

3]Te

chno

logy

−2.9

0−3

.30*

*−6

.69*

*6.

15**

[−1.

04]

[−2.

33]

[−3.

39]

[2.8

1]Fu

ndch

arac

teri

stic

sFu

ngag

e.2

5**

.04

.095

.91*

[2.4

6][.2

4][.5

9][2

.00]

Man

agem

ent

expe

nse

rati

o−3

.18

−4.3

0−6

.09

−3.8

8[−

1.47

][−

1.21

][−

1.56

][−

.06]

Log

(Tot

alA

sset

s)−1

.42*

**−1

.17

−1.3

6.1

1[−

3.19

][−

1.72

][−

1.44

][.1

5]R

etur

non

Ass

ets

(RO

A)

4.68

**.1

9.0

5−2

.66

[2.4

4][.0

6][.0

1][−

.87]

Mac

roec

onom

icva

riab

les

Bank

rate

−1.2

8**

−.95

*−.

92−.

32[−

2.70

][−

1.88

][−

1.14

][−

.93]

SP/

TSX

retu

rn2.

16.9

6−.

443.

54[1

.04]

[.46]

[−.1

8][1

.39]

Log

(Pro

vinc

ialG

DP

Gro

wth

)−2

4.89

−32.

88−3

3.84

−11.

10[−

.78]

[−.8

7][−

.65]

[−.4

3]

Obs

erva

tion

s24

423

823

821

024

121

014

862

R-s

quar

ed.0

8.0

8.1

1.1

0.0

8.1

4.1

8.3

5C

lust

erco

ntro

lYe

arPr

ovin

cean

dye

arPr

ovin

cean

dye

arPr

ovin

cean

dye

arYe

arYe

arYe

arYe

ar



TAB

LE

4C

onti

nued

Pane

lB:N

umbe

rof

priv

ate

firm

sin

the

port

folio

Mod

el1

Mod

el2

Mod

el3

Mod

el4

Mod

el5

Mod

el6

Mod

el7

Mod

el8

DID

DID

Inve

stee

char

acte

rist

ics

Fund

char

acte

rist

ics

Mac

roec

onom

icva

riab

les

All

vari

able

sO

ntar

iofu

ndon

lyO

ther

prov

inci

alfu

nds

Con

stan

t30

.75*

**31

.54*

34.9

6*−1

45.8

3**

24.6

7***

−139

.93*

**−8

7.53

***

−327

.97*

**[4

.57]

[2.4

0][2

.51]

[−2.

55]

[4.7

3][−

7.25

][−

6.00

][−

6.97

]Tr

eat

−15.

26**

−16.

06−1

4.45

−11.

94−1

5.98

**−1

0.78

***

[−2.

23]

[−1.

22]

[−1.

15]

[−1.

57]

[−2.

56]

[−3.

07]

Aft

er20

05−3

.70

−2.7

5−2

.31

−2.1

1−2

.82

−1.5

1−2

.72*

**−.

77[−

.54]

[−.2

3][−

.19]

[−.3

8][−

.45]

[−.3

8][−

3.04

][−

.17]

Tre

at*

Aft

er20

056.

415.

462.

82.3

84.

35−2

.88

[.92]

[.46]

[.23]

[.06]

[.66]

[−.6

9]In

vest

eech

arac

teri

stic

sE

arly

stag

e−1

0.66

**−6

.86*

**−5

.38*

**−7

.71*

**[−

2.70

][−

3.37

][−

4.23

][−

4.06

7]Te

chno

logy

−2.2

9−3

.36

.43

−5.9

8[−

.39]

[−1.

48]

[.17]

[−1.

01]

Fund

char

acte

rist

ics

Fung

age

.47

.32*

.54*

**−.

89[1

.69]

[1.8

8][3

.66]

[−.6

9]M

anag

emen

tex

pens

era

tio

27.0

3**

24.6

4***

13.1

0***

734.

38**

[2.5

9][6

.15]

[4.8

6][2

.54]

Log

(Tot

alA

sset

s)9.

85**

*9.

67**

*5.

79**

*19

.94*

**[3

.05]

[8.0

9][6

.40]

[8.9

6]R

etur

non

Ass

ets

(RO

A)

2.20

−.23

6.59

***

11.4

0[.3

4][−

.04]

[3.4

3][.6

0]M

acro

econ

omic

vari

able

sBa

nkra

te2.

69**

.92*

1.08

**2.

01**

[2.7

2][1

.76]

[2.2

4][2

.43]

SP/

TSX

retu

rn10

.60

2.12

−2.3

44.

17[1

.70]

[.73]

[−1.

66]

[.80]

Log

(Pro

vinc

ialG

DP

Gro

wth

)−1

16.4

3−8

7.56

*−8

.88

−396

.47*

**[−

1.74

][−

1.78

][−

.39]

[−3.

82]

Obs

erva

tion

s24

223

623

620

823

620

814

860

R-s

quar

ed.0

7.0

8.1

2.4

7.1

0.5

0.3

9.7

3C

lust

erco

ntro

lYe

arPr

ovin

cean

dye

arPr

ovin

cean

dye

arPr

ovin

cean

dye

arYe

arYe

arYe

arYe

ar

This

tabl

epr

esen

tsor

din

ary

leas

tsq

uare

(OL

S)pa

nel

regr

essi

ons

ofd

eter

min

ates

ofth

enu

mbe

rof

publ

icco

mpa

nies

(Pan

elA

)an

dnu

mbe

rpr

ivat

eco

mpa

nies

(Pan

elB

)in

the

port

folio

.St

anda

rder

rors

are

clus

tere

dby

year

for

Mod

els

1,5,

6,7,

and

8,an

dar

ecl

uste

red

bypr

ovin

cean

dye

arfo

rM

odel

s2,

3,an

d4.

Mod

els

1an

d2

pres

ent

the

dif

fere

nce-

in-d

iffe

renc

e(D

ID)

regr

essi

onre

sults

.Mod

els

3,4,

and

5pr

esen

tth

ere

sults

wit

hin

vest

eech

arac

teri

stic

s,fu

ndch

arac

teri

stic

s,an

dm

acro

econ

omic

vari

able

s,re

spec

tive

ly.M

odel

6pr

esen

tsth

ere

sults

wit

hal

lva

riab

les.

Mod

el7

pres

ents

the

resu

ltsw

ith

fund

sin

Ont

ario

only

and

mod

el8

pres

ents

the

resu

ltsw

ith

fund

sin

allo

ther

prov

ince

s.T-

stat

isti

csar

ein

squa

rebr

acke

ts.

*Sig

nific

ant

atth

e10

%le

vel

**Si

gnifi

cant

atth

e5%

leve

l**

*Sig

nific

ant

atth

e1%

leve

l



effect is significant at least at the 10 percent level – support-ing Hypotheses 1 and 2. In the most conservative estimate,after the tax policy change, there is an increase of 2.46 morepublic companies in the portfolios. The average number ofpublic companies in the fund portfolios is 4.16. After the taxpolicy change, there is a 59.13 percent (2.46/4.16 = 0.5913)increase in the number of public companies in the fundportfolios. Since the main objective of the LSVCCs is toprovide venture capital to private entrepreneurial firms,increasing investment in public companies crucially under-mines such objective. From a diversification perspective,individual investors expected their investment in a venturecapital fund in lieu of a mutual fund (a pooled investment inless liquid private companies rather than in the public com-panies invested in by mutual funds) to diversify their ownpersonal portfolios. This style drift instead may result ininvestors unknowingly taking on higher (unwanted) sys-tematic risk. LSVCCs may be becoming more mutual-likethan initially planned and expected. As a robustness check,we separated the funds between funds in Ontario and fundsin other provinces and retest Model 6. The results show thatthe changes in the tax policies in Ontario have significantimpact on LSVCC funds in Ontario but have no such impacton the funds in other provinces. These results may not besurprising as changes in the tax policies in Ontario shouldonly affect the individual investors in Ontario and notothers. In terms of numbers, we find that LSVCCs in Ontarioare more likely to double (4.14/4.16 = 0.9952) their numberof public companies in their portfolios. We do not observesuch effects for funds in other provinces. These results cor-respond well to our second hypothesis that after theannouncement of the elimination of the tax credit, LSVCCsin Ontario will increase their investments in public compa-nies, while LSVCCs in other province will not.

Other than the main variables of interest, our results alsofind that LSVCC funds that focus on early stage private firmsare less likely to invest in public companies. In terms ofprovinces, Ontario LSVCCs with early stage focus are lesslikely to invest in public companies; however, LSVCCs inother provinces with early stage focus are more likely toinvest in public companies. Both results are statistically sig-nificant at the 5 percent level. Similar to the early stage focus,Ontario LSVCCs with a focus on technology firms are lesslikely to invest in public firms but LSVCCs in other prov-inces with technology focus will be more likely to invest inpublic firms. We also find limited evidence that fund age,size and bank rate have an impact on the fund investment inthe public companies, but these results are not consistent.

Panel B of Table 4 illustrates the regression results for thenumber of private companies in the portfolio. In general, wecould not observe any effect on investment in the privatecompanies after the tax policy change announcement inOntario. When separating the funds into those in Ontarioand those in other provinces, we find that LSVCCs inOntario are more likely to reduce the number of privatecompanies in their fund portfolios. On average, LSVCCs inOntario have reduced the number of private companies intheir portfolios by 2.72. The average number is 20.66 and thisshows that after the tax policy change, there is a 13.17percent (2.72/20.66 = 0.1317) decrease in investment interms of the number of private companies. These results are

statistically significant at the 1 percent level. LSVCCs inother provinces have reduced the number of private compa-nies in their portfolios by 0.77. However, we are unable tofind any statistical significance.

Besides the policy variable we are interested in, the resultsin Panel B of Table 4 also show some other interesting find-ings. We find that LSVCCs with a focus on investment inearly stage firms are less likely to increase their investmentin private companies. We find that high managementexpense ratios (MER) are positively associated with invest-ment in private companies and larger funds are more likelyto increase investment in terms of the number of privatecompanies in the fund portfolio. Short-term interest rates arepositively associated with the number of private companiesin the portfolios as well. We find limited evidence that as thefunds grow older, the fund is more likely to invest in privatecompanies. However, these results are only statistically sig-nificant for funds in Ontario. A similar story applies to thesize of the funds.

So far, we have only tested the level of public and privatecompanies in the fund portfolios. As a robustness check, wealso test whether there is a change in the fund investmentallocation before and after the changes of tax policies inOntario in 2005. The results are illustrated in Table 5. Table 5provides the regression results on the percentage of totalinvestment in the public/private companies as well as inshort-term investments. The dependent variables are per-centage of total investment in public companies, percentageof total investment in private companies, and percentage oftotal investment in the short term. Standard errors are clus-tered by year for all models. Model 1 presents the results forLSVCC funds in Ontario only and Model 2 presents theresults for LSVCC funds in all other provinces.

In terms of percentage of total investment (measured inCanadian dollars) in public companies, LSVCCs in Ontarioare more likely to place heavier weight in public companiesand LSVCCs in other provinces are more likely to reduce theweight in public companies. Our results show that there isan increase of 6 percentage points as an investment increasein public companies for LSVCCs in Ontario. Considering theaverage percentage of total investment in public companiesis 12 percent, this means a 50 percent (0.06/0.12 = 0.5)increase in the weight of investment in public companies forLSVCCs in Ontario. There is also a reduction of 7 percentagepoints for investments in public companies for LSVCCs inother provinces. The average amount of public companiesis 12 percent and this represents a 58.33 percent (0.07/0.12 = 0.5833) reduction in the weight of the investment inpublic companies for LSVCCs in other provinces. Bothresults are statistically significant at the 10 percent level.Other than the main variable of interests, we also find thatLSVCCs in Ontario with a focus on early stage and technol-ogy are less likely to invest in public companies and LSVCCsin other provinces with a focus on early stage and technol-ogy are more likely to invest in public companies. Bank ratesare statistically significantly negatively related to the invest-ment in public companies for Ontario LSVCCs only. Fundage, stock market return, and local economic developmentare positively related to the percentage of total investment inthe public companies only for the funds that are outside theprovince of Ontario.



TA

BL

E5

Reg

ress

ion

Res

ult

son

the

Per

cen

tage

ofT

otal

Inve

stm

ent

inP

ub

lic/

Pri

vate

Com

pan

ies

asw

ell

asin

Sh

ort-

Ter

m/G

over

nm

ent

Bon

dIn

vest

men

ts

Perc

ent

ofto

tali

nves

tmen

tin

publ

icco

mpa

nies

Perc

ent

ofto

tali

nves

tmen

tin

priv

ate

firm

sPe

rcen

tof

tota

linv

estm

ent

insh

ort

term

/go

vern

men

tbo

nds

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

sO

ntar

iofu

ndon

lyO

ther

prov

inci

alfu

nds

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Con

stan

t.4

7*1.

04**

*−.

0001

3.5

3*−.

16[2

.04]

[3.5

7][−

.00]

[.45]

[1.7

0][−

.32]

Aft

er20

05.0

6*−.

07*

.07

.23*

*−.

13**

−.16

[1.9

5][−

1.94

][1

.53]

[2.5

7][−

2.51

][−

1.05

]In

vest

eech

arac

teri

stic

sE

arly

stag

e−.

09**

*.1

6***

.04

−.06

.05

−.10

[−4.

74]

[5.2

8][.6

2][−

1.23

][.5

7][−

.95]

Tech

nolo

gy−.

14**

*.2

4***

.12*

*−.

10.0

2−.

14[−

4.45

][4

.69]

[2.6

3][−

1.72

][.5

7][−

1.08

]Fu

ndch

arac

teri

stic

sFu

ngag

e.0

04.0

4**

.004

−.03

*−.

01−.

006

[1.1

2][2

.71]

[.53]

[−2.

16]

[−1.

17]

[−.2

7]M

anag

emen

tex

pens

era

tio

−.10

−3.6

3.0

7−1

.81

.03

5.45

[−1.

73]

[−1.

26]

[1.4

7][−

.35]

[.38]

[1.3

5]L

og(T

otal

Ass

ets)

−.01

−.07

***

.03*

*.0

5**

−.02

.02

[−1.

11]

[−4.

01]

[2.1

2][3

.06]

[−.9

4][.7

2]R

etur

non

Ass

ets

(RO

A)

.01

−.22

−.07

−.06

.06

.28*

**[.1

1][−

1.67

][−

.88]

[−.4

1][.7

2][3

.53]

Mac

roec

onom

icva

riab

les

Bank

rate

−.03

***

.004

−.02

−.04

*.0

4***

.03

[−3.

20]

[.26]

[−.8

9][−

1.81

][3

.01]

[1.0

5]SP

/T

SXre

turn

−.02

.14*

*.0

04−.

22*

.01

.08

[−.4

0][2

.26]

[.12]

[−1.

87]

[.20]

[1.1

6]L

og(P

rovi

ncia

lGD

PG

row

th)

−.42

2.77

**.7

5−1

.60

−.33

−1.1

7*[−

.71]

[3.1

2][1

.43]

[−.8

4][−

.74]

[−1.

97]

Obs

erva

tion

s14

862

148

6214

862

R-s

quar

ed.2

4.4

9.1

7.4

0.3

4.3

1C

lust

erco

ntro

lYe

arYe

arYe

arYe

arYe

arYe

ar

This

tabl

epr

esen

tsor

din

ary

leas

tsq

uare

(OL

S)pa

nelr

egre

ssio

nof

perc

enta

geof

tota

linv

estm

ent

inpu

blic

/pr

ivat

eco

mpa

nies

and

insh

ort-

term

inve

stm

ent.

Stan

dard

erro

rsar

ecl

uste

red

byye

ars

for

all

mod

els.

Mod

el1

pres

ents

the

resu

ltfo

rL

SVC

Cfu

nds

inO

ntar

ioon

lyan

dm

odel

2pr

esen

tsth

ere

sults

for

LSV

CC

fund

sin

othe

rpr

ovin

ces.

T-st

atis

tics

are

insq

uare

brac

kets

.*S

igni

fican

tat

the

10%

leve

l**

Sign

ifica

ntat

the

5%le

vel

***S

igni

fican

tat

the

1%le

vel



We find mixed results in terms of the percentage invest-ment in private companies and in terms of the percentageinvestment in short-term/government bonds (see Table 5).In general, after the announcement of the tax changes inOntario, we find that LSVCCs in other provinces are morelikely to invest in private companies but found no such dif-ferences for LSVCCs in Ontario. Ontario LSVCCs seem toshift investment in the short-term/government bonds topublic companies; however, we do not observe such shiftsfor LSVCCs in other provinces. So far, our empirical resultshave indicated that only LSVCCs in Ontario are shiftingaway from their original mandates to invest in entrepreneur-ial firms and to promote local economic growth, measuredeither by the number of and the percentage of private firmsin their portfolios which we interpret as clear support forHypothesis 3. Besides our main variable of interest, we alsofind that the fund size, as measured in terms of total assets,has a positive impact on levels of investment in private com-panies for LSVCCs both in Ontario and in other provinces.Similarly, past fund returns as measured in terms of ROAhave a positive impact on the short-term investment forLSVCCs in provinces outside Ontario. Also, LSVCCs inother provinces are less likely to put money into short-term/government bonds as markets improve. However, we couldnot find a similar effect for LSVCCs in Ontario.

ROBUSTNESS CHECK

On May 14, 2008, the Ontario government announced adelay to the phase-out of the tax credit by an additional year.In this section, we test whether the trend for OntarioLSVCCs to increasingly invest in public companies becameeven stronger after the announcement in 2008. Table 6 pro-vides an additional robustness check in relation to the trendfor Ontario LSVCCs to invest in public companies after thesecond announcement in 2008. The dependent variables arenumber of public and private companies in the portfolios,percentage of total investment in public/private companiesand short-term investments. Standard errors are clusteredby year for all models. Model 1 presents the results forLSVCC funds in Ontario only and Model 2 presents theresults for LSVCC funds in all other provinces.

Our results in Table 6 show that there is an increase of 8.15more public companies in the fund portfolios. In terms ofpercentage of total investment in public companies, we findthat there is a 10.5 percentage point increase. On the otherhand, LSVCCs in other provinces were more likely toincrease the percentage of total investment in private com-panies by 23.3. All those results are both economically andstatistically significant. The tendency for fund managers inOntario to invest in public companies strengthened evenmore with the imminent phase-out of the tax credits, whichis also in line with the predictions of our three hypotheses.

CONCLUSION

The primary purpose of this paper is to understand the dualroles played by government policy and the resulting ineffi-cient corporate governance structure of LSVCCs in the port-folio choices of LSVCCs in Canada, and especially the ones

situated in the province of Ontario. We know that govern-ment support has been found to be crucial for bridging thecapital gap faced by nascent entrepreneurial firms and forthe provision of venture capital (Chen, Lee, & Mintz, 2002;Cumming & Johan, 2012a; Di Giacomo, 2004; Laliberte,1998). However, without proper mechanisms and efficientgovernance structures, government-sponsored programsmay not be able to achieve such objectives. In this case, thegenerous tax subsidies initially provided to incentivize retailinvestors to invest in LSVCCs may have been misdirected asthey resulted in disincentives for the same retail investors todirect the appropriate corporate governance measuresagainst fund managers. The tax credits essentially mitigatedthe need for disparate investors to monitor fund managersas up-front benefits had already been obtained, and anyneed for returns postponed as such investments weremainly made with long-term benefits sought within retire-ment accounts. As LSVCCs have been found to be underper-forming even risk-free 30-day Treasury Bills (Cumming &Johan, 2012a; Cumming & MacIntosh, 2006, 2007), the usualdisciplining method used against other financial intermedi-aries could not be used as redemptions have been prohibitedas a result of lock-ups. As a result of lack of oversight orattention, the announcement of tax reform could only bringabout changes that may not necessarily be beneficial forinvestors. We show that the announcement of the impend-ing change in tax policy in Ontario in 2005 has resulted in astyle drift by the Ontario government-sponsored LSVCCs.LSVCCs in Ontario are more likely to include more publiccompanies in the fund portfolios and reduce their holdingsof private firms after the 2005 announcement, which not onlyreduces the diversification benefits of (retail) investors, butalso increases portfolio risk as well as other LSVCC risks asfund managers are experts in venture investment, notinvestment in public equities. Such drift, however, may benecessary to ensure increased liquidity as redemptions mayincrease pursuant to the lock-up periods expiring.

On the other hand, we find no evidence for changes ofweighting of public and private companies within the fundportfolios for LSVCCs in other provinces. Instead, theLSVCCs in other provinces are decreasing the percentage ofinvestments in public companies and increasing the percent-age of total investment to private companies within a similarperiod. We argue that two theoretical reasons could explainthis tendency of fund managers in Ontario to invest more inpublic companies. LSVCC managers in Ontario may initiatetournament behavior, drift away from their mandate toinvest in venture firms, and invest in more risky publiclylisted companies (measured by volatility), to counter or miti-gate the potential effect of the change in investor sentimentpursuant to the removal of the tax credit, and the potentialredemptions as the lock-up periods expire. Such risky styledrift may actually be facilitated by the lack of oversight byretail investors as they are not incentivized to monitor theirinvestments and the performance of fund managers. Albeitpaying LSVCC fund managers fees close to double those ofmutual fund managers, investors’ diversification benefitsare further eroded by the style drift. Increasing investmentin publicly listed companies results in an even widerfunding gap for entrepreneurs seeking venture capitalfunding. Extant research has found that retail-based venture



TA

BL

E6

Rob

ust

nes

sC

hec

ks

Num

ber

ofpu

blic

com

pani

esN

umbe

rof

priv

ate

firm

sPe

rcen

tof

tota

linv

estm

ent

inpu

blic

com

pani

esPe

rcen

tof

tota

lin

vest

men

tin

priv

ate

firm

sPe

rcen

tof

tota

linv

estm

ent

insh

ort

term

/go

vern

men

tbo

nds

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Mod

el1

Mod

el2

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Ont

ario

fund

only

Oth

erpr

ovin

cial

fund

s

Con

stan

t29

.76

−7.1

01−8

6.88

***

−334

.3**

*.4

28*

1.06

9***

.015

3−.

0159

.556

***

−.05

29[1

.77]

[−.4

4][−

6.13

][−

6.62

][1

.79]

[3.5

0][.0

5][−

.07]

[3.3

7][−

.12]

Aft

er20

088.

148*

**.1

67−3

.310

10.4

2.1

05**

*−.

0462

.051

7.2

33**

*−.

156*

*−.

187*

**[3

.62]

[.08]

[−1.

38]

[1.3

4][3

.15]

[−1.

13]

[.91]

[3.7

7][−

2.53

][−

4.45

]In

vest

eech

arac

teri

stic

sE

arly

stag

e−4

.660

***

2.01

6**

−5.3

70**

*−8

.385

***

−.08

62**

*.1

48**

*.0

334

−.01

82.0

527*

*−.

130*

*[−

3.33

][2

.91]

[−4.

19]

[−3.

40]

[−4.

39]

[4.5

1][1

.65]

[−.4

5][2

.33]

[−2.

44]

Tech

nolo

gy−6

.941

***

5.82

0**

.672

−7.7

55−.

148*

**.2

24**

*.1

12**

*−.

0738

.035

9*−.

150*

**[−

3.55

][2

.82]

[.26]

[−1.

30]

[−4.

90]

[4.4

3][3

.05]

[−1.

34]

[1.8

3][−

3.84

]Fu

ndch

arac

teri

stic

sFu

ngag

e.0

903

.727

**.4

96**

*−1

.427

.004

31.0

286*

*.0

0628

*−.

0071

4−.

0106

***

−.02

15**

[.91]

[2.5

4][3

.39]

[−1.

69]

[1.4

4][2

.18]

[1.8

4][−

.59]

[−3.

47]

[−2.

24]

Man

agem

ent

expe

nse

rati

o−5

.822

−1.8

2913

.25*

**80

5.9*

*−.

0939

−3.8

32.0

557

−.68

8.0

382

4.52

1[−

1.48

][−

.03]

[5.0

6][2

.54]

[−1.

68]

[−1.

28]

[.70]

[−.1

7][.9

7][.7

5]L

og(T

otal

Ass

ets)

−1.4

37.1

505.

860*

**19

.86*

**−.

0159

−.06

89**

*.0

292*

.035

4**

−.01

33.0

334

[−1.

47]

[.18]

[6.3

8][8

.80]

[−1.

18]

[−3.

80]

[1.8

3][2

.72]

[−1.

67]

[1.6

6]R

etur

non

Ass

ets

(RO

A)

−1.0

26−2

.933

7.21

0***

11.9

0−.

0040

3−.

233

−.08

52−.

0076

1.0

892

.240

[−.2

3][−

1.08

][3

.05]

[.70]

[−.0

4][−

1.77

][−

1.43

][−

.03]

[1.4

5][1

.16]

Mac

roec

onom

icva

riab

les

Bank

rate

.636

−.37

7.3

633.

847*

*−.

0044

1−.

0083

4−.

0034

1.0

207

.007

82−.

0124

[.77]

[−.8

2][.5

4][2

.18]

[−.4

3][−

.45]

[−.2

3][.7

6][.4

1][−

.61]

SP/

TSX

retu

rn6.

654*

*3.

221*

−5.5

57*

10.5

5.0

777*

.085

2.0

651

.015

9−.

143*

−.10

1[2

.71]

[1.9

4][−

1.80

][1

.32]

[2.1

1][1

.48]

[.85]

[.17]

[−2.

00]

[−1.

62]

Log

(Pro

vinc

ialG

DP

Gro

wth

)−1

4.64

−7.6

77−7

.688

−343

.9**

−.24

32.

667*

**.4

89−.

824

−.24

6−1

.842

[−.3

6][−

.25]

[−.2

2][−

2.73

][−

.47]

[3.0

8][.6

0][−

.77]

[−.2

5][−

1.16

]

Obs

erva

tion

s14

862

148

6014

862

148

6214

862

R-s

quar

ed.1

83.3

46.3

86.7

31.2

38.4

86.1

56.3

17.3

07.2

87C

lust

erco

ntro

lYe

arYe

arYe

arYe

arYe

arYe

arYe

arYe

arYe

arYe

ar

This

tabl

epr

esen

tsor

din

ary

leas

tsq

uare

(OL

S)pa

nelr

egre

ssio

nof

num

ber

and

perc

enta

geof

tota

linv

estm

ent

inpu

blic

/pr

ivat

eco

mpa

nies

and

insh

ort-

term

inve

stm

ent.

Stan

dard

erro

rsar

ecl

uste

red

byye

ars

for

all

mod

els.

Mod

el1

pres

ents

the

resu

ltfo

rL

SVC

Cfu

nds

inO

ntar

ioon

lyan

dm

odel

2pr

esen

tsth

ere

sults

for

LSV

CC

fund

sin

othe

rpr

ovin

ces.

T-st

atis

tics

are

insq

uare

brac

kets

.*S

igni

fican

tat

the

10%

leve

l**

Sign

ifica

ntat

the

5%le

vel

***S

igni

fican

tat

the

1%le

vel



capital funds set up with the best of intentions have failed toaccomplish what they were established for. We add to theliterature with our findings relating to the Canadian experi-ment and hope that our findings will encourage futureresearch to help create programs that will accomplish theirgoals.

ACKNOWLEDGEMENTS

The authors would like to thank Rajesh Chakrabarti,Douglas Cumming, Michele Meoli, Millan Jankovic and theseminar participants of the CGIR Special Issue Conferenceon “Global Perspectives on Entrepreneurship: Public andCorporate Governance” at the Schulich School of Business,York University, Toronto, April 25–26, 2013.

NOTES

1. However, unused tax credits cannot be carried over to thefuture. See http://www.fin.gov.on.ca/en/guides/lsif.html.

2. See http://www.fin.gov.on.ca/en/bulletins/lsif/0206.html.

REFERENCES

Barclay, M. J., Pearson, N. D., & Weisbach, M. S. 1998. Open-endmutual funds and capital-gains taxes. Journal of Financial Eco-nomics, 49: 3–43.

Bergstresser, D. & Poterba, J. 2002. Do after-tax returns affectmutual fund inflows? Journal of Financial Economics, 63: 381–414.

Brown, K. C., Harlow, W. V., & Starks, L. T. 1996. Of tournamentsand temptations: An analysis of managerial incentives in themutual fund industry. Journal of Finance, 51: 85–110.

Chaplinsky, S. & Haushalter, D. 2010. Financing under extremerisk: Contract terms and returns to private investments in publicequity. Review of Financial Studies, 23: 2789–2820.

Chen, D., Lee, F. C., & Mintz, J. 2002. Taxation, SMEs and entrepre-neurship. OECD Science, Technology and Industry WorkingPapers, No. 2002/09.

Chevalier, J. & Ellison, G. 1997. Risk taking by mutual funds as aresponse to incentives. Journal of Political Economy, 105: 1167–1200.

Cumming, D. 2006. The determinants of venture capitalportfolio size: Empirical evidence. Journal of Business, 79: 1083–1126.

Cumming, D. 2007. Government policy towards entrepreneurialfinance: Innovation investment funds. Journal of Business Ven-turing, 22: 193–235.

Cumming, D., Fleming, G., & Schwienbacher, A. 2008. Financialintermediaries, ownership structure and the provision ofventure capital to SMEs: Evidence from Japan. Small BusinessEconomics, 31: 59–92.

Cumming, D., Hass, L. H., & Schweizer, D. 2013. Private equitybenchmarks and portfolio optimization. Journal of Banking &Finance, 37: 3515–3528.

Cumming, D. J. & Johan, S. A. 2010. Phasing out an inefficientventure capital tax credit. Journal of Industry, Competition &Trade, 10: 227–252.

Cumming, D. J. & Johan, S. A. 2012a. Venture capital and privateequity contracting: An international perspective (secondedition). London: Elsevier Science Academic Press.

Cumming, D. J. & Johan, S. A. 2012b. Is venture capital in crisis?World Economic Review, July/August: 69–72.

Cumming, D. J. & MacIntosh, J. G. 2003a. A cross-country compari-son of full and partial venture capital exits. Journal of Banking &Finance, 27: 511–548.

Cumming, D. J. & MacIntosh, J. G. 2003b. Venture capital exits inCanada and the United States. University of Toronto LawJournal, 53: 101–199.

Cumming, D. J. & MacIntosh, J. G. 2003c. Canadian labor-sponsored venture capital corporations: bane or boon? In A.Ginsberg & I. Hasan (Eds.), New venture investment: Choicesand consequences: 169–200. London: Elsevier Science AcademicPress.

Cumming, D. J. & MacIntosh, J. G. 2005. Comparative venturecapital governance: Private versus labour sponsored venturecapital funds. Venture Capital, Entrepreneurship and PublicPolicy, 9: 69–94.

Cumming, D. J. & MacIntosh, J. G. 2006. Crowding out privateequity: Canadian evidence. Journal of Business Venturing, 21:569–609.

Cumming, D. J. & MacIntosh, J. G. 2007. Mutual funds that invest inprivate equity? An analysis of labour-sponsored investmentfunds. Cambridge Journal of Economics, 31: 445–487.

Del Guercio, D. & Tkac, P. A. 2002. The determinants of the flow offunds of managed portfolios: Mutual funds vs. pension funds.Journal of Financial and Quantitative Analysis, 37: 523–557.

Di Giacomo, M. 2004. Public support to entrepreneurial firms: Anassessment of the role of venture capital in the European expe-rience. Journal of Private Equity, 8: 22–38.

Gompers, P. A. & Lerner, J. 1999. The determinants of corporateventure capital success: Organizational structure, incentives, andcomplementarities. In R. K. Morck (Ed.), Concentrated corporateownership: 17–54. Chicago: University of Chicago Press for theNational Bureau of Economic Research.

Laliberte, P. 1998. The economic impact of labour-sponsoredventure capital corporations: A case study from Quebec. Inter-national Labour Review, 137: 47–60.

Leleux, B. & Surlemount, B. 2003. Public versus private venturecapital: Seeding or crowding out? A pan-European analysis.Journal of Business Venturing, 18: 81–104.

MacIntosh, J. G. 1997. Venture capital exits in Canada and theUnited States. In P. Halpern (Ed.), Financing growth in Canada:279–356. Calgary: University of Calgary Press.

Manigart, S., De Waele, K., Wright, M., Robbie, K., Desbrieres, P.,Sapienza, H., & Beekman, A. 2002. The determinants of therequired returns in venture capital investments: A five-countrystudy. Journal of Business Venturing, 17: 291–312.

Manigart, S., Korsgaard, M. A., Folger, R., Sapienza, H., & Baeyens,K. 2002. The impact of trust on private equity contracts. WorkingPaper, Vlerick Leuven Ghent Management School.

Osborne, D. & Sandler, D. 1998. A tax expenditure analysis oflabour-sponsored venture capital corporations. Canadian TaxJournal, 46: 499–574.

Sahlman, W. A. 1990. The structure and governance of venturecapital organizations. Journal of Financial Economics, 27: 473–521.

Sandler, D. 2004. Venture capital and tax incentives: A compara-tive study of Canada and the United States. Toronto: CanadianTax Foundation.

Sapienza, H. J., Manigart, S., & Vermeir, W. 1996. Venture capitalistgovernance and value added in four countries. Journal of Busi-ness Venturing, 11: 439–469.

Sirri, E. R. & Tufano, P. 1998. Costly search and mutual fund flows.Journal of Finance, 53: 1589–1622.

Wright, M. & Lockett, A. 2003. The structure and management ofalliances: Syndication in the venture capital industry. Journal ofManagement Studies, 40: 2073–2104.



Sofia Johan is with the Schulich School of Business, YorkUniversity and also an Extramural Research Fellow at theTilburg Law and Economics Centre (TILEC) in The Nether-lands. Her research is primarily focused on law and finance,sovereign wealth funds, market surveillance, hedge funds,venture capital, private equity, and IPOs. Her recent publi-cations have appeared in numerous journals including theAmerican Law and Economics Review, Journal of Financial Eco-nomics, Journal of Banking and Finance, European FinancialManagement, European Economic Review, and EntrepreneurshipTheory and Practice, among numerous other journals.

Denis Schweizer is an assistant professor of alternativeinvestments at WHU – Otto Beisheim School of Manage-ment. He studied business administration at the Johann-

Wolfgang-Goethe University in Frankfurt am Main,Germany. After graduating, Denis began his dissertation atthe European Business School (EBS) and completed his PhDin April 2008. From September 2011 until January 2012 hewas a visiting scholar at the New York University. DenisSchweizer has published in journals such as Journal ofBanking and Finance, Journal of Corporate Finance, and Euro-pean Financial Management, among others.

Feng Zhan is a PhD candidate in Finance at the SchulichSchool of Business, York University. He also holds Master’sDegrees in Economics from the University of Manitoba. Hisdissertation focuses on the efficiency and manipulationactivities across international equity markets.



Documents

The Changing Latitude: Labor-Sponsored Venture Capital Corporations in Canada