Mutual Funds that Invest in Pri

Document Sample
 Mutual Funds that Invest in Pri Powered By Docstoc
					                      Mutual Funds that Invest in Private Equity?

               An Analysis of Labour-Sponsored Investment Funds∗




                                              Douglas J. Cumming
                                                School of Business
                                               University of Alberta
                                                Edmonton, Alberta
                                                Canada T6G 2R6
                                               Tel: (780) 492-0678
                                               Fax: (780) 492-3325
                                      E-mail: Douglas.Cumming@ualberta.ca
                                      Http://www.bus.ualberta.ca/dcumming



                                             Jeffrey G. MacIntosh
                              Toronto Stock Exchange Professor of Capital Markets
                                                 Faculty of Law
                                             University of Toronto
                                                78 Queen’s Park
                                                Toronto, Ontario
                                               Canada M5S 2C5
                                              Tel: (416) 978-5785
                                              Fax: (416) 978-6020
                                        E-mail: j.macintosh@utoronto.ca



                                            First Draft: September 2001
                                               This Draft: June 2003



           ∗      We are grateful for comments from Mark Huson, Aditya Kaul, Janet Payne, Corrine Sellars, Wolfgang
Stummer and the seminar participants at the Canadian Law and Economics Association Annual Conference (Toronto, September
2001, 2002), Eastern Finance Association (Baltimore, April 2002), Academy of Entrepreneurial Finance and Business Ventures
Conference (New York, April 2002), Tilburg University Conference on Regulatory Competition (The Netherlands, September
2001), Northern Finance Association (Banff, September 2002), Financial Management Association (San Antonio, October 2002),
and the CESifo Conference on Venture Capital and Public Policy (Munich, November 2002). This paper is scheduled for
presentation at the 2003 American Law and Economics Association Annual Conference (Toronto, September 2003). The
Schulich School of Business National Research Program in Financial Services and Public Policy provided generous financial
support.
                    Mutual Funds that Invest in Private Equity?

            An Analysis of Labour-Sponsored Investment Funds



                                                  Abstract




        This paper considers the structure, governance and performance of a unique class of mutual funds
that receives capital only from individuals, and reinvests this contributed capital in private companies, as
opposed to traditional mutual funds that invest in publicly traded companies. We consider the particular
class of mutual funds known as Canadian Labour-Sponsored Investment Funds (LSIFs). We point out that
similar funds have been introduced in the UK; as well, it has been argued that the Canadian LSIF structure
should be introduced in the US. In contrast to expectations, we show that LSIFs have artificially low betas
(the average beta is 0.38), returns that have significantly underperformed industry benchmarks (including
risk-free 30-day T-bills), average MERs greater than 4%, and have collectively accumulated $CAN 8 billion
($US 5 billion) over 1992-2002. We show that these incongruous data are directly attributable to the LSIF
statutory structure. LSIF legislation mandates episodic valuations that determine share prices, an 8-year
investor lock-in period, and onerous constraints on capital reinvestment. LSIFs also afford to their investors
tax-generated returns in excess of 100%. The unique LSIF structure provides generalizable insights into the
relation between organizational governance and performance, and the unsuitability of mutual fund structures
for private equity investment.



Key words: Mutual Fund Performance, Labour-Sponsored, Private Equity, MER, Risk, Return, Fundraising

JEL classification: G23, G24, G28, G32, G38, K22
1. Introduction


         Mutual funds traditionally aggregate the capital contributions of individual investors and reinvest
their contributed capital mainly in publicly traded companies (see, e.g., Massa, 2003; Pastor and
Stambaugh, 2002; Wermers, 2000, Chen et al., 2000, Khorana, 2001, Carhart, 1997; Droms and Walker,
1996; Buttimer et al., 2001). Venture capital funds aggregate the contributions of institutional investors
(primarily pension funds), corporations, and wealthy individuals and invest these contributions in the
equity of private and potentially high growth technology firms (Sahlman, 1990; Berger and Udell, 1998;
Gompers and Lerner, 1999a; Schwienbacher, 2001; Neus and Walz, 2003; Hege et al., 2003). In some
countries, including Canada and the U.K., a new type of tax-advantaged mutual fund has emerged that is
essentially a hybrid between a traditional mutual fund and a traditional venture capital fund. This fund
structure has been attracting increasing attention; for example, it has been argued that the Canadian model
should be adopted in the US (see, e.g., Hebb, 1999).                 These funds collect the capital contributions of
individual investors, regardless of their net worth (as does a mutual fund), but invest in private equity (as
does a venture capital fund). The balance of portfolio firms ranges from investments in mostly traditional
sectors to mostly or entirely high tech. This paper examines one such fund: the Canadian Labour
Sponsored Investment Fund (LSIF). This paper relates the governance structure of this hybrid mutual
fund / venture capital fund to its performance.1


         At the highest level of generality, this paper engages two important issues. First, what is the linkage
between organizational structure and fund risk and return? Second, is this type of tax-advantaged vehicle a
target-efficient means of re-directing investment dollars into private entrepreneurial companies? In respect
of the first of these questions, we suggest that the unique organizational structure of the LSIF funds is a
primary reason for their extremely poor performance. In respect of the second, we suggest that the statutory
design of the LSIF fund programs is such as to frustrate, rather than contribute to achievement of their
primary objective.


   1
         In related work, Martin and Petty (1983) analyze the performance of publicly traded venture capital companies. Kleim
(1998) considers the case of mutual funds that invest in small-cap stocks; see also Nanda and Singh (1998). Previous research on
LSIFs has considered a fairly narrow range of issues, including governmental tax expenditures of LSIFs, the performance of
LSIF entrepreneurial firms, and the impact of LSIFs on the market for private equity. In particular, Osbourne and Sandler (1998)
compute the direct costs of LSIFs to the Canadian taxpayers (see also Vaillancourt, 1997), Halpern (1997), Brander et al. (2002)
show that LSIF entrepreneurial firm investments underperform entrepreneurial investments from other Canadian private equity
investments (see Anderson and Tian, 2003, for a Monte Carlo experiment) and Cumming and MacIntosh (2002) present evidence
that LSIFs have crowded out traditional venture capital funds.
                                                                                                        2

        LSIF data present a puzzling picture. Since LSIFs invest in comparatively risky small (and
private) entrepreneurial firms, we would naturally expect to observe average betas in excess of one, in
addition to risk-adjusted returns in excess of standard market benchmarks (Smith and Smith, 2000;
Cochrane, 2001; Kerins et al., 2003; Astebro, 2003). Surprisingly, the data indicate the exact opposite:
almost all LSIFs have betas less than 1 (with an average 3-year beta of 0.378), and LSIF returns are
consistently lower than all industry-wide benchmarks over a 10-year horizon, including 30-day risk-free t-
bills. Despite gross underperformance, LSIF MERs are as high as 8%, and the average MER is 4.03%
(nearly twice as high as the average Canadian mutual fund MER). Equally astonishing, given their poor
performance and high MERs, is that over a 10-year period LSIF assets under administration have grown
from less than $1 billion to more than $8 billion.


        The analysis of LSIF betas, returns and asset accumulation presents some rather unique
challenges. LSIF share prices are not determined in the market, but by periodic evaluations of net asset
values per share as determined by the board of directors (for interim reporting periods) and by an
independent valuer (for year-end reporting), with some variation in the frequency of these valuations.
Therefore, LSIF returns are not driven by CAPM-type assumptions and/or the Fama and French (1993,
1995) factors, etc. For this reason, we refer to LSIF betas as “pseudo-betas”. These pseudo-betas are not
an accurate measure of systematic risk, but at best constitute a measure of the relative risk across the
different LSIFs. As well, because share prices are determined by periodic valuations, they are not
amenable to time-series analysis. We thus analyze LSIF returns using a cross-sectional analysis over 1-
month, 3-month, 6-month, 1-year, 3-year, and 5-year horizons. In the cross-sectional analysis, we show
LSIF pseudo-betas are inversely related to returns (in contrast to the standard positively sloped security
market line), over measurement horizons up (and including) to 5-years.


        We suggest that the extremely low LSIF returns are a product of statutory constraints that
inefficiently constrain managerial choice and investment choice. In particular (depending on the province
of incorporation), LSIFs must reinvest up to 80% of their contributed capital within a 1 to 3-year period
of the contribution date, and pay severe penalties for non-compliance. This constraint can adversely
affect returns in two ways – first, by forcing managers to commit capital to inferior investment projects;
and second, by attenuating the due diligence process that lies at the heart of equity investing (Gompers
and Lerner, 1999a) and adversely impacting the quality of the deal forged with various of a LSIF’s
portfolio companies. Further statutory constraints exacerbate this problem by limiting the allowable types
of investments in entrepreneurial firms, imposing constraints on the structuring of investments and the
                                                                                                           3

size of an investment in any given firm, and limiting the geographical situs of investee firms. In addition,
in contrast to their private venture fund counterparts, LSIF funds must be structured as corporations,
sacrificing the various advantages associated with limited partnership form.


         We show that, like private venture capital funds (Gompers and Lerner, 1999b), LSIF management
remuneration consists of a combination of fixed fees (ostensibly to cover out-of-pocket costs) and carried
interest (i.e., a percentage of the appreciation in the value of the fund’s investment portfolio, typically
paid after certain performance hurdles are met). Unlike either mutual funds or venture capital funds,
however, LSIF funds typically split the functions performed by the manager between the “investment
manager” and the “advisor”. The former either invests the funds’ assets, or advises the fund’s board of
directors on the investment of assets, and the latter performs a variety of other administrative and
marketing functions. Both LSIF managers and advisors are typically well compensated compared to both
mutual funds that invest in publicly-traded securities and private venture capital funds that invest in
private equity (with an average MER, again, in excess of 4%). We argue that the statutory generated
dearth of competition among LSIFs likely accounts for the excessively high compensation to LSIF
managers and advisors. In a cross-sectional analysis across the 50 LSIFs, we find higher fixed fees are
associated with lower fund risk and lower fund returns. Higher carried interest fees tend to be associated
with higher fund risk and returns, as well as greater capital accumulation. Hence, even in the unusual
case of LSIFs investing in private companies, incentive fees do matter in mutual fund governance and
performance, just as in the case of mutual funds that invest in public companies (see, e.g., Elton et al.,
2001).


         Finally, despite extremely low LSIF returns, we show that a significant amount of capital has
being contributed to LSIFs. Despite earning rates of return lower than most bond indices, by the end of
2001 LSIFs had collectively accumulated CAN$8.4 billion (US$ 5.5 billion) of capital.                Capital
contributions have been sufficiently robust that, in the face of an inability to invest contributed capital
within statutory constraints, Canada’s two largest LSIF funds have both at one time or another suspended
or limited capital contributions. Perhaps most surprising of all, we show that LSIF asset accumulation is
negatively related to comparative rates of return among funds that have been in existence for more than 1
year, and unrelated to comparative rates of return among funds that have been in existence for more than
3 years (in contrast to evidence relating to mutual funds that invest in publicly traded firms; see e.g. Baks
et al., 2001; Chevalier and Ellison, 1997, 1999a,b). This surprising result casts grave doubt on the
wisdom of the provincial and federal governments’ LSIF programs
                                                                                                                                 4


         While we focus on a unique institutional setting, we nonetheless believe that our results have
general implications for the efficacy of focused government subsidization of venture capital. We note
that venture capital organizations similar to Canadian LSIF have been introduced in other countries, such
as the U.K.2 In response to the perceived importance of venture capital to the funding of entrepreneurial
firms, many governments have mounted programs that seek to foster venture capital financing. Such
programs have been the subject of previous scholarly examination (Cressy, 2002; DeMeza, 2002; Lerner,
1999, 2002; Gompers and Lerner, 2001; Kanniainen and Keuschnigg, 2000, 2001; Keuschnigg, 2002;
Keuschnigg and Nielsen, 2001, 2002). Prior work, however, has not considered the financial market
properties of tax subsidized mutual funds that invest in private equity.


         This paper is organized as follows. Section 2 explains in detail the LSIF corporate governance
mechanisms. Tax incentives to invest in LSIFs are explicitly detailed in section 3. Sections 4 and 5
describe the LSIF data pertaining to risk, betas, returns, compensation, and fundraising. The determinants
of LSIF betas, returns and asset accumulation are evaluated in section 6, 7, and 8, respectively. Section 9
discusses limitations and future research. Concluding remarks follow.


2. LSIF Mandate, Structure, and Governance


         Elsewhere, we explore the identity of LSIF investors, in addition to LSIF mandates, structure,
governance, and taxation (Cumming and MacIntosh, 2002). In this section, we briefly summarize this and
related research on both LSIF funds and private Canadian funds.


         A LSIF fund may be incorporated in any province that has passed legislation specifically allowing
for the creation of a LSIF. It may also be incorporated under similar federal legislation, in which case it may
operate in any province that has passed legislation specifically authorizing federal LSIFs to operate in that
jurisdiction. In either case, it must have a labour union sponsor, whose only participation in practice is to
“rent” its name to the fund in return for a small portion of net assets, or a fixed annual fee. Only individuals
may invest in LSIFs,3 and minimum investment requirements are modest (typically less than or equal to


  2
         Venture Capital Trusts (VCTs) were introduced in the U.K. in 1995. Cumming (2003) notes UK VCTs are similar in
structure to Canadian Labour Sponsored Investment Funds (Cumming’s discussion paper is based on our paper).
  3
          While precise data is not available, it would appear that, in Canada, corporations are the largest contributors to private
                                                                                                                                      5

$1000). Most LSIF contributions are made through the vehicle of a registered retirement savings plan
(“RRSP”, which roughly corresponds with a 401k plan in the United States). As further outlined below, all
LSIF investments up to a certain amount (typically $5,000) receive generous federal and provincial tax
credits, and contributions made through an RRSP receive additional tax benefits. In most provinces, LFIF
investors are locked into their investments for eight years. Premature withdrawal results in repayment of the
federal and provincial tax credits, and typically also results in the fund levying a withdrawal fee on the order
of 6% of the investor’s withdrawal amount.


           LSIF legislation typically specifies multiple fund objectives, including regional development,
enhancing financing for small firms, creating jobs, furthering worker education, and in some cases advancing
the cause of unionised enterprise. Many of the LSIF funds, however, have stated that they principal or even
sole objective is the pursuit of profits (although in Quebec, non-profit objectives are pursued with some
vigour).


           LSIFs legislation requires that LSIFs be formed as corporations, in contrast to the limited partnership
form usually favoured by private funds. Mandatory use of the corporate form potentially gives rise to a
number of adverse consequences. In particular, the use of corporate form sacrifices the discipline that the
limited life span of a partnership imposes on management. The use of corporate form also diminishes
contractual flexibility in constructing the nexus of contracts that underlies LSIF operation. While in theory
use of the corporate form imposes a more exacting disciplinary structure on LSIF management than we
observe in the case of private limited partnership (since shareholders directly elect the directors of the fund),
the atomization of share ownership sacrifices most if not all of these benefits, since collective action and free
rider problems ensure that few if any shareholders have the appropriate incentives to monitor or discipline
fund manages. Moreover, LSIF legislation invariably requires that the sponsoring labour union receive a
class of shares which, while not participating in dividends or assets on winding up, is entitled to appoint a
majority of the directors.         Few unions have shown any interest in exercising their powers of control, and
many have contractually delegated their power to appoint directors to the management company that is
contractually engaged to manage the fund. This results in a pure separation of ownership from fund control.




funds, followed by pension funds and individuals. The balance between these different types of investors varies substantially from one
year to the next. This contrasts somewhat with the situation in the United States, in which pension funds are the largest contributors of
capital, followed by corporations and individuals (Gompers and Lerner, 1998, 1999a; Smith and Smith, 2000).
                                                                                                             6

        Further exacerbating the shareholders lack of control, many LSIFs contract with external parties for
the supply of vital functions such as investment management. Because purely contractual duties are not as
easily policed or enforced as duties performed internally under a command and control system, this
contracting out has the effect of creating additional slack in the investor-manager relationship.


        LSIFs are subject to a number of statutory constraints (see Osbourne and Sandler, 1998, for
particulars), including: (1) all LSIFs are constrained to invest in the sponsoring jurisdiction (as
determined by the location of the investee firm’s assets, employment, or other similar factors), (2) an 8-
year investor lock-in period, (3) restrictions on the number of allowable funds in certain jurisdictions, (4)
statutory penalties for failure to reinvest fixed percentages of contributed capital in private entrepreneurial
firms within a stated period of time (typically 1 to 3 years, and (5) constraints on the size and nature of
investment in any given entrepreneurial firm.           These constraints can have the effect of forcing
investments to be made in inferior firms and/or without adequate due diligence, limiting competition
across LSIFs, and limiting investor discipline through threat of withdrawal of capital contributions.
These constraints are discussed in more detail below in conjunction with the analysis of the effect on
betas, risk, returns and asset accumulation.


        Many covenants suitably designed to mitigate opportunistic behavior among limited partnership
venture funds (Gompers and Lerner, 1996, 1999a) are absent among LSIFs, including restrictions on the
use of debt, restrictions on co-investment by the organizations earlier or later funds, restrictions on
coinvestment by VC managers (general partners), restrictions on fundraising by VC managers, and
restrictions on other actions of VC managers, among other things.


        In short, there is reason to believe, from a theoretical perspective, that LSIF funds are an inferior
organizational form.    The separation of ownership from control and the absence of other meaningful
constraints on management, coupled with the presence of statutory limitations not present for either mutual
funds or private venture capital funds, are highly suggestive that LSIF rates of return will lag those on both
mutual funds and private venture capital funds.


3. The Tax Position of LSIF Investors


        Many individuals contributing to LSIFs have no other investments (Vaillancourt, 1997), and thus
risk serious underdiversification. Given their inefficient structure, and the risks and illiquidity associated
                                                                                                                             7

with private equity investments, why would underdiversified investors ever contribute their savings to the
LSIF class of mutual funds? The answer lies in the federal provincial and federal tax relies given to LSIF
investors. Combining the tax credit and tax deductibility features of LSIF investments, and (depending on
the tax bracket of the individual investor), the after-tax cost of a $5,000 LSIF investment made through the
vehicle of an RRSP (see section 2) ranges from $1180 to $2390, or roughly 27 to 48 percent of the nominal
dollar cost of the investment. The governmental sponsors effectively pay the balance of the cost. An
individual investor holding for the required hold period (typically eight years4) will reap a return on
investment in excess of 100%, even if the LSIF generates no profits at all (Osbourne and Sandler, 1998,
provide complete tax details, and a comparison that shows LSIF investors receive a considerable tax
advantage relative to other types of funds in Canada).


          Investors are concerned about after-tax, rather than before-tax returns (Bergstresser and Poterba,
2000, re mutual funds; see also Barclay et al., 1998, re open-ended mutual funds and capital gains taxes).
The tax subsidies accorded LSIF investors thus result in LSIFs having a substantially lower required rate of
return than either mutual funds or private venture capital funds (Cumming and MacIntosh, 2003).


          The tax-expenditure cost of LSIFs to the various Canadian governments is extremely large.
Combined federal and provincial tax credit expenditures for the period 1992-2002 were approximately
$3.27 billion.5 This amounts to 33.4% of all contributions made to LSIFs during this period of time. It is
likely that this understates the total tax cost, insofar as there is evidence (Vaillancourt, 1997) that most
LSIF investors invest through tax-deductible RRSPs, and that a non-trivial number of these investors
would not otherwise have made an RRSP investment. For these investors, the additional tax expenditures
associated with the deductibility of their RRSP contributions also count as a relevant tax cost.


4. LSIF Capital Under Administration and the “Overhang” of Uninvested Capital


          LSIFs have accumulated more capital that the aggregate of all other types of private equity funds in
Canada (including private limited partnerships, corporate funds, etc.). By the end of 2001, LSIFs had

   4
          Early withdrawal of contributed funds makes all tax credits repayable. In addition, most funds have levy a fee on the
order of 6% of the individual’s holding for early withdrawal.
   5
          This figure was computed from various annual “Tax Expenditures and Evaluations” reports for the pertinent years. See
e.g. Tax Expenditures and Evaluations 2002 (Ottawa: Her Majesty the Queen in Right of Canada (Dept. of Finance), 2002) (Cat.
No. F1-27/2002E), Table 2, at http://www.fin.gc.ca/taxexp/2002/taxexp02_e.pdf.
                                                                                                                            8

accumulated CAN$8.4 billion (US$5.5 billion) of capital under management (in 2001 dollars). Figure 1
indicates the growth of LSIF capital over the 1992 – 2001 period (the years for which the Canadian Venture
Capital Association (CVCA) has reported this information6).


                                                 [FIGURE 1 ABOUT HERE]


          Figure 2 presents CVCA data for aggregate capital under management in the venture capital
industry, capital available for investment, and new capital contributions in each year over 1992-2001.
“Capital available for investment” indicates funds that have been contributed but not invested; it reflects
the extent to which contributions to venture capital funds have outstripped the funds’ ability to invest
these contributions. It can be seen from Figure 2 that, historically, there has been a large “overhang” of
uninvested capital in Canada, much of which has accumulated in the LSIFs. In fact, CVCA data
underestimates the overhang by approximately 30%.7 By the end of 1996, the overhang amounted to
approximately three years of venture capital investments (Department of Finance (Canada), 1996).                         The
problem of overhang forced one of Canada’s largest LSIFs (Working Ventures) to suspend new capital
raising for two and a half years (from mid-1996 to the end of 1998). At the time of suspension, Working
Ventures had only 19% of its contributed capital invested in eligible businesses.8 More recently, because it
failed to meet its statutory requirement to have invested 60% of capital contributions in qualifying
businesses, Solidarity (the largest LSIF) has limited capital contributions for the 2002-2003 year.9


                                                 [FIGURE 2 ABOUT HERE]


          Because of this problem of “overhang”, LSIF legislation now typically requires LSIFs to invest their
capital contributions within 1-3 years following receipt (or to have invested a certain proportion of their
capital in eligible entrepreneurial ventures by year’s end). If an investment deadline looms, this can force the
fund to make improvident investment choices in order to avoid severe penalties.



   6
          See www.cvca.ca and www.canadavc.com.
   7
          The CVCA data are based on the faulty assumption that LSIF legislation requires that certain percentages (20-40%) of
contributed capital not be invested in entrepreneurial firms. This amount of capital is removed from the CVCA’s overhang
calculation. In fact, the legislation merely limits the types of instruments that non-invested capital be invested in.
   8
          See "Working Ventures Puts Capital Raising on Hold" www.newswire.ca...June996/05/c0564.html
   9
          See Fonds de Solidarite Annual Report, 2002, available at http://www.sedar.com/.
                                                                                                                                   9

          The tax-driven structure of the LSIFs tends to exacerbate the problem of overhang. In contrast to
contributions to private funds, most contributions to LSIFs are made through RRSPs (see sections 2 and 3),
with the consequence that the vast proportion of contributions are made in the final three months of the tax
year (February-April). Moreover, unlike private funds, in order to secure the tax benefits associated with
investing in an LSIF, investors must commit their funds up front, rather than making contractual
commitments that are subsequently drawn as needed. This makes LSIF fundraising extremely “lumpy”
(again worsening the overhang problem).


5. LSIF Characteristics and Performance: Univariate Data Analysis


          Figure 3 presents the performance of LSIFs over the 1992 – 2003 period. Figure 3 clearly indicates
that a LSIF index has underperformed comparable indices, even a 30-day risk-free bond index.10 This
underperformance is generally consistent with recently developed theoretical work (Kanniainen and
Keuschnigg, 2001; Keuschnigg, 2002; Keuschnigg and Nielsen, 2001, 2003a,b; Lerner, 2002) on the effect
on the venture capital market of providing tax subsidies to a subset of venture funds (Lerner, 1999, provides
seminal empirical work). That LSIFs have seriously underperformed other types of funds while attracting
substantially more capital (Figure 1) strongly suggests an inefficient allocation of capital.11


                                                [FIGURE 3 ABOUT HERE]


          Table 1 presents summary statistics for the gross returns and the various other characteristics of each
of the 50 LSIFs (note that no LSIF has ever been wound up, so that there is no issue of survivorship bias).
The reported “beta” (which is calculated only for funds with 3 of more years of returns and reported by Globe
Funds www.globefund.com) is not a real beta, in the sense that it does not reflect market-driven price

   10
          Canadian data sources for Figure 3: www.globefund.com, www.morningstar.ca. The data do not exhibit survivorship
bias because there has not been an LSIF that has been wound up (the tax benefits provided to these funds tend to ensure that
capital inflows will occur regardless of performance). The US VC Index value from Peng (2001) is not available for 2000 and
2001.     Peng’s    data   are   from   Venture    Economics.       Venture    Economics       has   posted   on   their   web   page
(www.ventureeconomics.com) a value of their own index for the date 06/28/2002 (only) of 361.36 that is based over a similar
horizon used by Peng. The authors owe thanks to Peng for directing us to the Venture Economics cite for a recent comparable
value for the US index. It is noteworthy that Peng’s index calculations are more economically and statistically rigorous than that
posted by Venture Economics.
   11
          Moreover, Cumming and MacIntosh (2002) show that LSIFs have crowded out other forms of private equity
investment in Canada, further exacerbating the problem of inefficient allocation of capital.
                                                                                                                         10

volatility. Rather, it reflects the volatility of assessments of value made periodically (usually quarterly) by the
board of directors, and at year’s end by nominally independent valuers. We argue below that uncertainty in
valuing private companies, the infrequency of LSIF reassessments of value, and perverse incentives of the
fund and its valuer all conspire to make valuations “sticky”, resulting in an artificial lowering of reported
“betas”. The betas are thus most useful in comparing the relative riskiness of different types of funds, rather
than as absolute measures of systematic risk. We discuss this issue further in section 6 below.


           In addition, unlike mutual funds or private venture capital funds, LSIFs have both managers and
advisors. Unfortunately, there is no perfectly consistent definition of either manager or advisor, nor are
reported managerial or advisory charges broken down between the various functions that a manager or
advisor might perform. In order to be as consistent in our classification as possible, we designated an
entity as the “manager” (no matter what the designation in the prospectus) if it handled investments or
advised the board on investments (although a “manager” may also perform advisory services). The
advisor category includes entities performing other administrative functions (such as cash administration
and other back office functions, and in some cases marketing).


           A number of stylised facts, many puzzling but addressed below in the subsequent sections, are
observed from the univariate presentation and analysis of the data in Tables 1:12
       •   LSIF betas average 0.378. This is unexpectedly low, given that LSIFs invest a significant part of
           their portfolios (60-80%, if statutory targets are met) in comparatively high risk private
           entrepreneurial companies (cf. Smith and Smith, 2000; Cochrane, 2001; Kerins et al., 2003;
           Astebro, 2003).
       •   There is negative relationship between LSIF beta and fund return, in contrast to the standard
           positively sloped security market line.
       •   LSIFs with greater proportions of their portfolios invested in equity (as opposed to bonds or cash)
           have not achieved superior returns, in contrast to the premium typically associated with equity
           (note that the bond and equity proportions used herein are the typical holdings by the fund over a
           3-year horizon, as reported by www.globefund.com).
       •   LSIF performance has not improved with age. This runs counter to both theory and evidence
           suggesting that the greater the experience of private venture fund managers, the greater the return



  12
           A correlation matrix on file with the authors indicates similar univariate patterns as reported in Table 1.
                                                                                                                           11

            (e.g., Chan et al., 1990; Bergmann and Hege, 1998; Gompers and Lerner, 1999; Kanniainen and
            Keuschnigg, 2001; Keuschnigg, 2002; Keuschnigg and Nielsen, 2001, 2003a,b).
        •   LSIF MERs are very high relative to other mutual funds. The sum of the fixed fees paid to
            managers and to advisors averages 3.45%, which is only slightly less than the average MER of
            4.03% in 2001.13 By contrast, the average MER for all Canadian mutual funds is 2.29% (and
            1.50% in the US; see Ruckman, 2003).
        •   Despite the fact that LSIF funds have had returns that grossly lag pertinent market indices (and
            mutual funds, whose returns roughly track the broad market index), and in the face of extremely
            high MERs, LSIFs have collectively attracted approximately $8 billion from investors over the
            1992-2002 period.


                                                [TABLE 1 ABOUT HERE]


            The following sections explore the puzzling aspects of the LSIF data in a multivariate analysis of
betas (section 6), returns (section 7), and asset accumulation (section 8). The ordering of the analysis of
these interrelated issues in the following sections is based on causality running from risk to returns to
fundraising (causal relations that are consistent with asset pricing theory and with prior literature on
mutual funds, as discussed further below; endogeneity tests did not indicate the presence of potential
bias).


6. LSIF Betas


6.1. Model Specification


            To understand cross-sectional variation across different LSIF betas, it is important to understand that
LSIF betas (which average 0.378; see Table 1) are artificially low, for three reasons. First, valuations are not
made continuously, as in the public market, but only at periodic intervals (typically monthly and/or quarterly,
and yearly). Interim (i.e., quarterly) valuations are typically made by the board of directors on the advice
of a valuation sub-committee. Legislation requires that an independent valuer make year-end valuations.
Episodic reporting of values produces a downward bias in betas (Fowler et al., 1979). Second, valuations of

   13
            This indicates that most of the fees earned by LSIF managers are paid in the form of fixed fees rather than carried
interest.
                                                                                                                                12

private companies (and especially technology firms and firms in the early stages of development) are made in
the face of extreme uncertainty, and are thus subject to large confidence internals. For this reason, unlike
firms in the public markets, revaluations are typically made only in the face of large changes in the firm’s
circumstances (again biasing “beta” downwards).                   Third, LSIF managers have an incentive to smooth
portfolio company valuations in order to reduce apparent risk, facilitating marketing efforts to retail investors.
The key importance of low risk in marketing efforts is suggested by even a casual scrutiny of LSIF web sites
(which, in almost every case, base their marketing efforts on low risk and generous tax credits, rather than
returns14). LSIF web pages and mutual fund reporting services routinely fail to explain to investors that LSIF
betas are artificially low.15


          The risk aversion of LSIF managers provides a supporting incentive to smooth the valuation stream.
LSIF managers collect most of their remuneration in the form of fixed fees (Table 1). They therefore have an
incentive to limit the extent of revaluations in order to smooth reported net asset values, maintaining a steady
collection of fixed fees, and ultimately the managers’ personal income streams. While in theory the
independent valuer acts as a brake on the extent to which management can ‘fix’ valuations to effect this end,
the fund managers have virtually unfettered control over the identity of the fund’s valuer. Because of LSIF
organizational structure, managerial choices are virtually immune from challenge by investors, who lack the
power of control. While control resides in the Labour fund sponsor, its compensation takes the form of either
a fixed annual fee or a small percentage of net asset value (typically 0.25%). Where it collects a portion of
net asset value, its incentive is clearly to maximize net asset value rather than profitability. Even in cases in
which it collects a fixed fee, its only interest is to ensure the survival of the fund – which (particularly since
the managers are fixed fee collectors) is better served by maximizing net asset value than by high returns.
While the natural assumption (for which there is empirical support in the case of mutual funds that invest in

   14
          See e.g. “Not Just a Pretty Tax Credit” at http://www.crocusfund.com/advisor/printconcept14.html. In its promotional
material, the Crocus Fund indicates that in the period following September 11, 2001, its returns have fallen less than the public
market. This is highly misleading in at least two respects.     First, as indicated earlier, LSIF companies are valued not by the
market, but by the fund (interim reporting periods) and an independent valuator (year end reporting period). Because of the
inherent difficulty of valuing private companies and the skewed incentives of the board and the independent valuer, valuations
tend to be “sticky” and subject to less fluctuation than public market prices, regardless of the underlying fundamentals.
Moreover, the web site fails to present information relating to returns for periods prior to September 11, 2001, during which the
Crocus Fund significantly underperformed the market (see Table 1 (Manitoba)).
   15
          Despite intensive searches over a 2-year period, the authors still have yet to find a single source that explains why LSIF
betas are artificially low. We may infer that a casual or unsophisticated investor is likely to infer that widely reported LSIF betas
accurately represent fund risk. More generally, see e.g., Elton and Gruber (2000), and Elton et al. (2002).
                                                                                                                             13

public companies) is that fund profitability and the ability to attract capital contributions are closely related,
our empirical analysis of asset accumulation (below) suggests a virtually complete disjunction between
annual returns and fund raising for LSIFs. In short, the Labour union has little incentive to question
management’s selection of a valuer; in fact, it shares management’s incentive to reduce apparent risk in the
interest of attracting capital contributions. With virtually unchecked management power to appoint a valuer,
the nominally independent valuer is thus closely dependent upon the goodwill of management, and likely to
produce valuations that are congenial to management. For these reasons, we suggest that reported “pseudo-
betas” are more useful as an indicator of the comparative than the absolute systematic risk of LSIFs.


            A further outgrowth of the periodic nature of LSIF valuations is that we cannot consider risk and
returns in time series regressions based on daily price change; rather, we are limited to examining only cross-
sectional differences in risk and returns across the 50 LSIFs. In the regressions that follow, we present a
variety of alternative cross-sectional specifications to illustrate robustness (other specifications, which are
available on request, did not yield materially different results). Our cross-sectional regressions across the 50
LSIFs account for the following factors:
        •   The percentage of equity and bond holdings in entrepreneurial firms (we would expect higher equity
            holdings to be associated with more risk and higher betas; see, e.g., Smith and Smith, 2000;
            Cochrane, 2001; Kerins et al., 2003; Astebro, 2003),16
        •   LSIF age, and a dummy variable for organizations with more than 1 LSIF (we would expect these
            factors to proxy for fund experience, which could impact fund risk and betas),
        •   Fixed fees and carried interest percentages (since both fixed fees and carried interest to may have an
            impact on managerial risk taking and fund beta),
        •   Dummy variables for different jurisdictions allowing for the creation of a LSIF (a dummy for
            Ontario is excluded to avoid perfect collinearity), and
        •   Dummy variables for funds that invest primarily in early stage entrepreneurial firms (a dummy for
            funds without a stage focus is suppressed), technology sector entrepreneurial firms across the
            spectrum of technology sectors, and specific sectors of technology firms (a dummy for funds without
            a sectoral focus is suppressed).


   16
            Equity, bond and cash holdings are based on the www.globefund.com reports of the typical equity holdings (etc.) by
the fund. As indicated in Table 1, this information is not reported for some funds. For the purpose of the regressions, for the
funds where this information was not available, we used the average of the other LSIF equity (etc.) holdings for those funds. The
regression estimates are quite robust to alternative specifications (available upon request).
                                                                                                         14


6.2. Regression Estimates


        Table 2 provides regressions for the fund pseudo-betas (Models (1) - (4)) and return variance
(Models (5) - (8)) as recorded in the GlobeFund databank (based on calculations for funds with at least 3
years of data). There is a surprisingly weak link between the percentage of equity in a fund’s portfolio
and both pseudo-beta and return variance. In only one of three models in Table 2 (Model (4)) in which
equity holdings were included as an independent variable were we able to find a statistically significant
link between equity holdings and pseudo-beta, and the coefficient is small (a 10% increase in equity
holdings increases beta by 0.04). Moreover, in none of our regressions was there a statistically significant
link between the percentage of bond holdings and pseudo-beta. The weakness of the link may be
attributable to variations in valuation practice, which introduce noise into pseudo-beta calculations. If so,
this casts serious doubt on the intercomparability of LSIF portfolio valuations. We did, however, find a
statistical link between equity holdings and returns variance in two of three regressions (Models (7) and
(8) indicate a 10% increase in equity holdings increases returns variance by approximately 1.4%).


        Table 2 also indicates a number of regional variations in pseudo-betas: Quebec and Manitoba
have lower pseudo-betas and returns variance, while the federally incorporated funds have higher pseudo
betas and returns variance.


                                        [TABLE 2 ABOUT HERE]


        Older funds also have higher pseudo-betas (Models (2) and (3): an increase in age by 1 year is
suggestive of an increase in beta by 0.003). There is also some evidence that older funds have greater
returns variance (Model (7): an increase in age by 1 year increases returns variance by 0.06%), although
statistical significance was achieved in only 1 of 3 models.         Overall, the data suggest that more
experienced LSIF managers undertake investments in riskier entrepreneurial firms, although our results
are not robust to all specifications.


        We would expect higher carried interest fees to be associated with higher fund risk, since
managerial carried interest fees are, in essence, managerial call options, and heightened volatility
increases the value of these options. We indeed find that a 10% increase in carried interest is associated
with an increase in beta of 0.2, and an increase in returns variance by 8.7%. Conversely, we would expect
                                                                                                                               15

that higher fixed fees will produce lower fund risk, since higher fixed fees will induce risk averse
managers to reduce the volatility of fund returns in order to increase the present value of their personal
income stream. We also find that this is the case; a 1% increase in fixed fees is associated with a
reduction in beta by 0.2, and lowering of returns variance by 6.1%.


           We would have expected generalist funds (which invest in both technology and non-technology
sectors) to have the lowest betas, followed by general technology funds, and specific technology funds.
The more general the investment focus, the greater the degree of diversification and the lower the degree
of asset specificity (which is generally associated with enhance risk). We would also have expected funds
that invest in relatively risky early stage ventures to have comparatively high betas. However, the data
are not consistent with these expectations. The general technology focus dummy in Table 2 is in fact
positive and significant in respect of both betas and returns variance, with coefficients on the order of
0.3). However, the early stage and specific technology dummies are negative and significant (early stage
betas are lower by 0.5; specific tech fund betas are lower by 0.3). We believe that the result in respect of
early stage ventures is attributable to inherent difficulties in valuing early stage enterprises. While early
stage ventures may have a product prototype in place, many such ventures will yet to have generated
revenues, let alone profits. Because of this, the tendency of the fund’s board of directors and its
independent valuers is to avoid re-valuations (and to carry investments at historical cost), lowering
apparent volatility.


7. LSIF Returns


7.1. Model Specification


           Similar to our analysis of pseudo-betas, in this section we present cross-sectional regressions for the
determinants of fund returns. The independent variables that we include are:
       •   The percentage of equity and bond holdings in entrepreneurial firms (we would expect a premium
           associated with higher equity holdings),
       •   LSIF pseudo beta,
       •   Fund age, and a dummy variable for organizations that market more than 1 LSIF (we would expect
           both of these factors to proxy for fund experience, which can in turn affect returns),17

  17
           Pseudo-betas are included in the regressions to proxy for fund risk. Recall that for reasons discussed above, this is not
                                                                                                                                16

        •   Fixed fees and carried interest percentages (we would expect lower fixed fees and higher carried
            interest to be associated with greater managerial incentives and higher returns),
        •   Dummy variables for general technology funds that invest across the spectrum of technology sectors,
            funds that focus specifically on early stage firms (a dummy for funds without a stage focus is
            suppressed), and funds with a specific technology focus (a dummy for funds without a sector focus is
            suppressed).
        •   Dummy variables for the LSIF’s fee structure, including dummies for no load funds and funds with
            back end loads18 (since loads have empirically been shown to be associated with the degree of
            competition between funds, and the presence of market power will affect a fund’s ability to earn
            economic rents: a dummy for front end fees is suppressed)
        •   Dummy variables for initial minimum and subsequent minimum RRSP purchase levels (which also
            proxy for the degree of market power possessed by a fund) and,
        •   Dummy variables for various jurisdictions (a dummy for Ontario is excluded to avoid perfect
            collinearity; the rationale for these variables is also discussed immediately below).


            The jurisdiction dummy variables are important for the following reasons. Different LSIFs operate
in different competitive environments, given that the number of LSIFs incorporated in each province varies
dramatically. LSIFs registered in different provinces do not compete with one another: residents in a
province can only purchase shares of a LSIF located in their province of residence (which must in turn make
investments in entrepreneurial firms based in that province). Thus, in Ontario, for example, investors wishing
to invest in a LSIF may choose between 43 funds (including the 29 funds incorporated in Ontario and 14
funds incorporated federally), allowing investors a wide range of choice. By contrast, in Manitoba, investors
may only invest in 2 LSIFs. While federally incorporated LSIFs are allowed to operate in Saskatchewan,
Ontario, New Brunswick, Nova Scotia and Prince Edward Island, most focus their marketing activities on
Ontario; some do not even maintain offices outside of Ontario. This heightens the extent to which funds in
the smaller provinces (in which few funds are typically incorporated) may exercise market power.


            One view is that funds in non-competitive environments will earn a lower rate of return than other


an actual beta. Diagnostic tests did not indicate problems associated with endogeneity. Note, however, that we do not include
assets under administration, as this is (potentially) endogenous to returns. See section 7 below.
   18
            Loads charges (fees charged to investors for the privilege of investing) are either charged on the date of the investment
– a “front-end load” – or upon exiting the investment – a “back end load”.
                                                                                                                              17

LSIFs, because the managers of such funds face less pressure than those of other funds to manage their assets
effectively. However, in competitive jurisdictions, there is likely to be a great deal more competition for
deals with entrepreneurial firms, putting upward pressure on deal prices and reducing returns (cf. Gompers
and Lerner, 2000). It is thus not entirely clear what effect the presence or absence of competition (as proxied
by load charges and minimum purchase levels: see section 2) will have on a fund’s return. Nonetheless, it is
plausible – indeed probable - that the degree of competitiveness of the environment in which a fund operates
will affect fund returns.


7.2. Regression Estimates


          Table 3 provides cross-sectional regressions for the determinants of returns across LSIFs for the 1-
month, 3-month, 6-month, 1-year, 3-year and 5-year periods. The results are robust for a variety of other
specifications not presented but available upon request.


                                                [TABLE 3 ABOUT HERE]


          One of the more puzzling results in Table 3 is the consistently negative (and significant)
relationship between fund returns and beta. This does not appear to be an artifact of the technology
“bubble” of 1999-2002, since it is consistent across all time horizons, including the five-year horizon.19
Thus, for example, funds with betas that were 0.5 higher than the average had annualized returns that
were approximately 0.05% to 0.1% lower than the average (the greatest coefficients are for the 1 year
horizon ending June 2002).


          Moreover, in the shorter performance periods (up to six months), funds with a greater proportion
of equity in their portfolios experienced statistically lower returns (a 10% increase in equity holdings is
associated with a 0.01% lowering in annualized returns), although in longer performance periods (one to
five years) there was no statistically significant relationship between equity holdings and returns. It may
be that the poorer performance over the shorter horizons reflects the economic downturn of 2001-2002.
Similarly, for bond holdings and for horizons up to 3 months, a 10% increase in bond holdings is
   19
          For funds with less than 3-years of data for the returns regressions in models (1) – (4) in Table 4 (for horizons of less
than 3 years) beta was specified as beta the average LSIF beta of 0.378 (see Table 1). Various alternative specifications did not
materially affect the estimates. As well, note that the estimates for the 3- and 5-year horizons are quite similar, and for those
regressions no assumptions had to be made as betas were available for all funds in those regressions.
                                                                                                      18

associated with greater annualized returns by 0.01%. However, over longer measurement intervals, the
percentage of bond holdings was not statistically related to returns. Overall, these results suggest that
LSIF management is at best value-neutral (if not value-destructive) with respect to its equity investing
activities.


         Over investment horizons up to and including one year, older LSIFs experienced lower returns
than younger funds (over the one year period, returns were lowered by 0.0006%). In the three and five
year performance periods, older funds did not experience either higher or lower returns. That LSIF
returns for older funds are either lower or at best no higher than other funds refutes one of the most
common defenses of the LSIF programs among practitioners and policy makers. A common refrain
begins with the observation that in venture capital investing, “the lemons ripen quickly, while the plums
ripen slowly” (e.g., Gompers and Lerner, 1999a). The apology concludes with the assertion that the LSIF
programs are still too young for the plums to have been brought to fruition. However, the data do not
indicate superior performance among the older funds.


         With respect to fixed fees, we would expect that funds that charge higher fixed fees will have a
greater incentive to purchase conservative investments in order to generate a reasonably constant stream
of fixed fees and reduce the variance of managers’ remuneration. This will reduce returns. Over longer
time horizons (one, three and five year returns) there appears to be no systematic relationship between
fixed fees and returns. Over shorter horizons up to 1 year, higher manager fixed fees appear to be
associated with lower returns. Curiously, however, higher advisor fixed fees are associated with higher
returns (although these results are sensitive to the econometric specification).


         The function of carried interest is to align manager interests with those of shareholders, and to
motivate managers to produce profits (Gompers and Lerner, 1999b; Elton et al., 2001). On this basis,
higher carried interest payments should be associated with higher returns. There is evidence that this is
the case over the three and five year time horizons in Table 3 (Models (9), (11) and (12) (a 10% increase
in carried interest is associated with greater annualized returns of approximately 0.03%). Over the time
horizons of up to one year (Models (1) – (8) in Table 4), there is a negative relation between carried
interest and returns. Because the one year horizon spans the 2001-2002 downturn in the economy, the
lower returns among funds with higher carried interest charges suggests that funds with higher carried
interest have invested in riskier entrepreneurial firms.
                                                                                                           19

        We would generally expect risk to increase for funds that focus either on a relatively narrow part
of the industrial spectrum (i.e. general technology and specific technology funds) or on early stage firms.
Consequently, standard asset pricing theory suggests that these funds will exhibit higher returns. The
evidence is supportive for general technology and specific technology funds, but not for early stage funds.


        Funds with back end fees and no loads had statistically superior returns, but only over short
performance horizons (one, three and six months). There is no evidence that LSIFs with higher minimum
initial and subsequent RRSP purchase levels produced higher returns. In contrast, James and Karceski
(2001), show (for mutual funds that invest in public companies) that higher minimum investment
requirements are associated with enhanced performance.


        In sum, a few puzzling findings stand out from the regression on the cross-sectional determinants
of LSIF returns. First, there is the systematically negative relationship between pseudo-betas and fund
returns over all horizons. Second, older funds do not produce higher returns. This evidence runs contrary
to U.S. evidence suggesting that older private venture capital funds have experienced higher returns. It
also belies one of the most persistent defenses of the LSIF programs – that insufficient time has elapsed to
conduct a meaningful review of LSIF profitability.


        These two findings (along with in Figure 3 showing underperformance) are suggestive of low
managerial skill. Alternatively, they evidence the effect of the statutory constraints noted earlier on
managerial behavior.


8. LSIF Asset Accumulation


8.1. Model Specification


        Similar to the prior specifications, in this section we present cross-section regressions for the
determinants of assets under administration. The variables considered include:
    •   LSIF age, and a dummy for organizations with more than 1 LSIF (we expect these factors to be
        associated with greater accumulation of assets, due, respectively, to greater time to accumulate assets
        and heightened investor recognition),
    •   Dummy variables for back end fees and no load fees (a dummy for front end fees is suppressed), and
                                                                                                                          20

            for the minimum RRSP and subsequent RRSP purchase levels (since fee structures could naturally
            be expected to affect investor incentives to invest),
        •   Fixed fees and carried interest percentages (we would expect lower fixed fees and higher carried
            interest to be associated with a greater alignment of investor and manager incentives, resulting in
            greater accumulation of assets under administration),
        •   Dummy variables for LSIFs that focus on early stage entrepreneurial firms (a dummy for funds
            without a stage focus is suppressed), general technology sector entrepreneurial firms, and specific
            technology entrepreneurial firms (a dummy for funds without a sector focus is suppressed) (we
            include these variables to account for sectoral investment preferences among investors),
        •   Dummy variables for the various jurisdictions sponsoring LSIF funds (a dummy for Ontario is
            excluded to avoid perfect collinearity), and
        •   LSIF returns since fund inception (other horizons are not considered simultaneously to avoid
            collinearity). There is strong evidence that, in general, fund performance is a strong determinant of
            assets under administration (in the mutual fund literature, see, e.g., Baks et al., 2001; Chevalier and
            Ellison, 1997, 1999a,b; in the venture capital literature, see Gompers and Lerner, 1998).


8.2. Regression Estimates


            Table 4 presents regressions on the cross-sectional determinants of LSIF assets under
administration for funds that have been in existence for at least 1 year, 3 years and 5 years.20 Perhaps the
most interesting result in Table 4 is that for LSIFs the quantum of assets under administration is unrelated
to fund rates of return for the funds that have been in existence for more than 3- and 5-years, and
negatively related to fund rates of return for the funds that have been in existence for more than 1 year.
These results are robust to a variety of other specifications of the explanatory variables not presented
(e.g., returns over different horizons, etc., which are not reported together due to collinearity issues), and
other specifications of the dependent variable not presented for reason of succinctness.21 The lack of a

   20
            Note that Quebec’s Solidarity fund, the oldest LSIF, is an outlier with $4.5 billion in assets. As such, we also
considered regressions without the Solidarity fund; for those regressions, the coefficients were identical (but the model
diagnostics are affected). Likewise, none of the estimates in the regressions in Tables 2 - 4 were affected by the presence of
Quebec’s Solidarity fund (and/or any other particular fund); additional specifications are available upon request.
   21
            For example, specifications (available upon request) with the dependent variable measured as changes in capital
accumulation over different points in time did not show any significant relation to LSIF rates of return. The specification
presented is the change in capital accumulation from time of fund inception to the period ending 6/1/2002.
                                                                                                        21

relationship (and negative relationship for the specification including the nascent funds) between returns
and asset growth is very potent evidence of the importance of tax incentives in inducing investment into
LSIFs. As well, the evidence is suggestive of a lack of sophistication of LSIF investors; even tax-favored
investors should not be indifferent to comparative rates of return when deciding where to invest.


        The lack of a relationship between fund returns and asset accumulation may also be a product of
statutory constraints placed both on investors that limit investor choice. As noted earlier (see sections 2
and 3), for provincially incorporated funds, provincial tax credits are available only to residents of the
province in which the fund is incorporated. Moreover, once a LSIF investment is made, LSIF investors
are locked into their investments for a period of 8 years. Early withdrawal results not only in a penalty
fee levied by the fund (which is typically 6% of the individual’s stake), but in the retroactive loss of all
tax credits (see note 4 and accompanying text). This severely constraints investor mobility once an
investment is made, and hence the ability to switch from a less profitable to a more profitable fund. The
regulatory structure therefore effectively drives a wedge between returns and asset accumulation across
different LSIFs. We also note that LSIFs typically market their product not on the basis of a complete
picture of fund returns (see notes 13 and 14), but on the generous tax benefits available, artificially
generated claims of low risk, and selectively presented returns data. We believe that this also accounts, in
part, for the lack of a relation between returns and asset accumulation.


                                        [TABLE 4 ABOUT HERE]


        Table 4 contains further evidence that tax incentives drive asset accumulation. If tax savings are
the primary determinant of LSIF investing, then fund age should be a predominant statistically significant
predictor of asset growth. Table 4 suggests that this is the case; fund age is associated with asset growth
over all three performance horizons (one, three and five years). Moreover, the economic importance of
the age variable increases in the regressions for the older funds: for the larger sample including funds of
more than 1, 3, and 5 years in age, each extra year is associated with approximately an extra $2 million,
$3 million, and $4.5 million in assets, respectively. We also note that greater asset accumulation took
place over a five-year horizon if the fund was part of a fund group (more than 1 LSIF per fund, which is
associated with approximately $146 million extra in assets).         This result reinforces the view that
economies of scale exist in fund marketing and in achieving enhanced investor recognition (see, e.g.,
Dermine and Roller, 1992).
                                                                                                         22

        Finally, we note that fee structures are significant determinants of capital accumulation. Funds
with higher managerial carried interest have attracted more capital (the economic significance of the
effect varies across the specifications, but generally a 10% higher carried interest percentage is associated
with $50 - $70 million more in assets). Overall, this evidence suggests that investors believe that higher
carried interest charges will yield higher returns. This expectation is in line with extant theory and
evidence concerning private venture capital funds, and suggests that LSIF investors are not completely
unsophisticated.   This rational behavior among LSIF investors suggests that statutory constraints
inhibiting competition between funds have accounted for the lack of a relationship between fund returns
and asset accumulation. A lack of competition between funds is also consistent with our observation that
LSIF management fees are generally well in excess of those charged either by mutual funds that invest in
public companies, or by private venture funds.


9. Limitations and Future Research


        Our cross-sectional regression analyses are primarily limited by the degrees of freedom available
in our data, in that there are only 50 LSIFs, and many have not been in existence for very long. As
discussed, time series analyses were not appropriate, as LSIF returns and betas (etc.) are based on
periodic portfolio valuations that determine fund price (as reported by Globe Funds www.globefund.com)
and there is no fund market price in the traditional sense. As noted earlier, some have suggested that
because many of the LSIFs have been formed in the past three or four years, it is premature to evaluate
performance. Our results do not support this criticism. LSIF funds have underperformed 30-day risk free
T-bills over almost all of the 1992 – 2002 period, and older funds do not have higher returns.
Nonetheless, it will obviously be instructive to re-evaluate LSIF performance at points in the future.


10. Summary and Concluding Remarks


        Labour Sponsored Investment Funds (LSIFs) were created with the intention of promoting
investment in small and medium-sized entrepreneurial firms, with an emphasis on the technology sectors.
The mechanism for promoting such investment was to create what is essentially a hybrid between a
mutual fund (capitalized by retail investors and investing mainly in public companies), and a venture
capital fund (capitalized by institutional investors, corporations and high net worth individuals, and
investing primarily in private entrepreneurial companies). Only retail investors may invest in a LSIF, and
minimum contributions are low, facilitating the capture of investments from low net worth individuals.
                                                                                                          23

Investors are generously subsidized via tax credits and, when the investment is made through an RRSP
retirement vehicle, deductibility from income. Aggregate tax expenditures on LSIF tax credits (but not
deductibility) for 1992-2002 were roughly $3.27 billion, amounting to 33.4% of all contributions made to
LSIFs during this period of time.


         An examination of LSIFs serves as an object lesson in the effects of organizational design on firm
performance. The governance structure of LSIFs is established not by the market, but by statute. This
statutory structure sacrifices many of the organizational advantages of the private venture capital limited
partnership. It also drives a wedge between ownership and control, creating sub-optimal managerial
incentives. It also imposes statutory constraints on the investment activities of LSIFs that are not shared
either by mutual funds or by private venture capital funds, and directs LSIFs to pursue goals other than
profit maximization (which, however, appear to be observed only by a small minority of funds). Further
statutory constraints limit investor choice at the time when the initial investment is made by conferring
tax credits only on investors who are resident in the province of incorporation. Moreover, the eight year
mandatory hold period, coupled with the recapture of tax credits for premature withdrawal, make it
impossible as a practical matter for investors to shift their capital between competing funds once the
investment has been made. These statutory constraints severely limit the extent of competition between
LSIF funds, allowing them to charge management fees that are well in excess of those charged by either
mutual funds or private venture capital funds. These high fees, coupled with the inferior returns of LSIF
shareholders suggest that LSIF managers are the primary, if not the exclusive beneficiaries of the
government tax subsidies.


         The gross return of the LSIF index over the 1992 – 1999 period was 28%, compared to 160% for the
TSE 300 Index, 180% for the Globe Canadian Small Cap Peer Index, and 650% for the US Venture Capital
Index (as computed by Peng, 2001). While LSIF betas are extremely low (averaging 0.378), consistent with
returns that lag a broad market index, we suggest that there are a number of reason why these “pseudo-betas”
cannot be taken as a true measure of LSIF risk. These include the fact that valuations are not conducted in an
open market, are made infrequently, and are distorted by managerial (and valuer) incentives to understate
portfolio risk.


         The evidence presented in this paper is at odds with LSIF managerial learning, whereby managers
at older funds develop their skills and generate superior returns. In our data, despite having higher
pseudo-betas, older funds produced lower returns. This counters a common defence of LSIFs, which is
                                                                                                          24

that insufficient time has passed to evaluate the performance of LSIFs, many of which are recent entrants
to the field and have thus not had time to generate returns over a full investment cycle. The inability of
more experienced managers to generate superior returns may also be reflective of the vast capital inflows
into LSIF funds (generated by the generous tax incentives), coupled with statutory constraints that not
only limit the geographical scope (and type) of investments that may be considered, but force managers to
invest contributed capital within mandated time periods ranging from one to three years of receipt.


        Surprisingly, we found strong evidence that fund pseudo-betas are inversely related to returns
(unlike the standard positively sloped security market line), over all horizons up to 5-years. Allied with
evidence suggesting that funds that invest more in entrepreneurial equity do not generate higher returns, this
suggests that LSIF managers are do not exhibit a high degree of skill in selecting entrepreneurial investees.
This calls into question the primary motivation behind the creation of the LSIF programs, which has been
to foster equity investment in fledgling entrepreneurial enterprises.


        Despite their low returns compared to other asset classes, and MERs averaging over 4% (and as
high as 8%), LSIFs have attracted aggregate capital inflows on the order of $8 billion in 10 years.
Perhaps the most surprising empirical result in our study is that there is no statistically verifiable
relationship between particular LSIF returns and asset accumulation for LSIFs that have been in existence
for more than 3 years, contrary to widespread evidence relating to mutual funds that invest in public
companies. In fact, including all funds that have been in existence for only 1 year, there is a negative
relationship between fund returns and LSIF asset accumulation. This may be contrasted to evidence
suggesting that asset accumulation by mutual funds that invest in public companies is closely tethered to
fund performance. We feel confident in suggesting that, but for their tax generated returns, LSIFs would
not exist.


        It has been strenuously argued that the Canadian LSIF model should be adopted in the US in order to
further stimulate investment in entrepreneurial companies (see, e.g., Hebb, 1999).             Similarly, the
policymakers are considering expanding tax concessions to U.K. Venture Capital Trusts, which have a very
similar structure to Canadian LSIFs (Cumming, 2003). The Canadian experience with LSIFs indicates that
similar structures should not be adopted or fostered in other countries.
                                                                                                   25

References

Astebro, T. (2003). “The Return to Independent Invention: Evidence of Unrealistic Optimism, Risk
       Seeking or Skewness Loving?” Economic Journal 113, 226-239.

Anderson, S., and Y.S. Tian (2003). “Incentive Fees, Valuation and Performance of Labour Sponsored
       Investment Funds.” Working Paper, Ryerson University.

Baks, K., A. Metrick, and J.A. Wachter (2001). “Should Investors Avoid All Actively Managed Mutual
       Funds? A Study in Bayesian Performance Evaluation.” Journal of Finance 56, 45-86.

Barclay, M.J., N.D. Pearson, and M.S. Weisbach (1998). “Open-End Mutual Funds and Capital-Gains
        Taxes.” Journal of Financial Economics 49, 3-43.

Berger and Udell (1998). “The Economics of Small Business Finance: The Roles of Private Equity and Debt
        Markets in the Financial Growth Cycle.” Journal of Banking and Finance 22, 613-673.

Berger, A.N., and G.F. Udell (2002). “Small Business Credit Availability and Relationship Lending: The
        Importance of Bank Organizational Structure.” Economic Journal 112, F32-F53.

Bergmann, D., and U. Hege (1998). “Venture Capital Financing, Moral Hazard, and Learning.” Journal of
      Banking & Finance 22, 703-735.

Bergstresser, D.B., and J.M. Poterba (2000). “Do After-Tax Returns Affect Mutual Fund Inflows?”
        Working Paper, Massachusetts Institute of Technology.

Brander, J.A., R. Amit, and W. Antweiler (2002). “Venture Capital Syndication: Improved Venture
       Selection versus the Value-Added Hypothesis.” Journal of Economics and Management Strategy
       11, 423-452.

Buttimer, R.J., D.C. Hyland and A.B. Saunders (2001). “IPOs and Mutual Fund Returns.” Working
       Paper, University of Texas at Arlington, University of Texas at Arlington, and Ohio State
       University.

Carhart, M.M. (1997). “On Persistence in Mutual Fund Performance” Journal of Finance 52, 57-82.

Carpenter, R.E., and B.C. Petersen (2002). “Capital Market Imperfections, High-Tech Investment, and New
       Equity Investment.” Economic Journal, 112, F54-F72.

Chan, Y-S, D.R. Siegel, and A.V. Thakor (1990). “Learning, Corporate Control and Performance
      Requirements in Venture Capital Contracts.” International Economic Review 31, 365-382.

Chen, H.-L., and Jegadeesh, and R. Wermers (2000) “The Value of Active Mutual Fund Management: An
       Examination of the Stockholdings and Trades of Fund Managers” Journal of Financial and
       Quantative Analysis, 343-368.

Chevalier, J. and G. Ellison (1997). “Risk Taking by Mutual Funds as a Response to Incentives” Journal
       of Political Economy 114, 389-432.
                                                                                                   26

Chevalier, J. and G. Ellison (1999a). “Career Concerns of Mutual Fund Managers.” Quarterly Journal of
       Economics 105, 1167-1200.

Chevalier, J. and G. Ellison (1999b). “Are Some Mutual Fund Managers Better than Others? Cross-
       Sectional Patterns in Behavior and Performance.” Journal of Finance 54 875-899.

Cressy, R. (1996). “Are Business Startups Debt-Rationed?” Economic Journal 106, 1253-1270.

Cressy, R. (2002). “Funding Gaps: A Symposium.” Economic Journal 112, F1-F16.

Cumming, D.J. (2003). “The Structure, Governance and Performance of U.K. Venture Capital Trusts,”
      Journal of Corporate Law Studies, forthcoming.

Cumming, D.J. and J.G. MacIntosh (2002). “Crowding Out Private Equity: Canadian Evidence” Working
      Paper, University of Alberta and University of Toronto.

Cochrane, J. (2001). “The Risk and Return of Venture Capital.” National Bureau of Economic Research
       Working Paper No. 8066.

DeMeza, D. (2002). “Overlending?” Economic Journal 112, F17-F31.

Dermine, J., and L.-H. Roller (1992). “Economies of Scale and Scope in French Mutual Funds.” Journal of
       Financial Intermediation 2, 83-93.

Droms, W.G., and D.A. Walker (1996). “Mutual Fund Investment Performance.” Quarterly Review of
       Economics and Finance 36, 347-363.

Elton, E., M.J. Gruber (2000). “The Rationality of Asset Allocation Recommendations.” Journal of
        Financial and Quantitative Analysis 35, 27-41.

Elton, E., M.J. Gruber, and C.R. Blake (2001). “Incentive Fees and Mutual Funds.” Journal of Finance,
        forthcoming.

Elton, E., M.J. Gruber, and J.A. Busse (2002). “Are Investors Rational? Choices Among Index Funds.”
        Journal of Finance, forthcoming.

Fama, E.F., and K.R. French (1993). “Common Risk Factors in the Returns on Stocks and Bonds.” Journal
       of Financial Economics 33, 3-56.

Fama, E.F., and K.R. French (1995). “Size and Book-to-Market Factors in Earnings and Returns.” Journal
       of Finance 50, 131-156.

Fowler, D.J., H.C. Rourke, V.M. Jog (1979). “Heteroscedasticity, R-square and Thin Trading on the
        Toronto Stock Exchange.” Journal of Finance 34, 1201-1210.

Gompers, P.A., and J. Lerner (1996). “The Use of Covenants: An Empirical Analysis of Venture
      Partnership Agreements.” Journal of Law and Economics 39, 463-98.
                                                                                                      27

Gompers, P.A., and J. Lerner (1998). “What Drives Venture Fundraising?” Brookings Proceedings on
      Economic Activity – Microeconomics, 149-192.

Gompers, P.A. and J. Lerner (1999a). The Venture Capital Cycle. Cambridge: MIT Press.

Gompers, P.A., and J. Lerner (1999b). “An Analysis of Compensation in the U.S. Venture Capital
      Partnership,” Journal of Financial Economics 51, 3-44.

Gompers, P.A., and J. Lerner (2000). “Money Chasing Deals? The Impact of Fund Inflows on the Valuation
      of Private Equity Investments.” Journal of Financial Economics 55, 281-325.

Halpern, P. (1997). Financing Growth in Canada, (editor) University of Calgary Press.

Hebb, T. (1999). “Canadian Labour Sponsored Investment Funds: A Model for U.S. Economically
       Targeted Investments.” Paper presented at The Second National Heartland Labour-Capital
       Conference, April 29-30, Washington, D.C.        A draft of this paper is available at
       http://www.heartlandnetwork.org/conference4_99/downloads/Hebb.pdf

Hege, U., F. Palomino and A. Schwienbacher (2003). “Determinants of Venture Capital Performance:
       Europe and the United States.” Working Paper, HEC School of Management, Tilburg University
       and the University of Amsterdam.

James, C., and J. Karceski (2001). “Captured Money? Differences in the Performance Characteristics of
       Retail and Institutional Mutual Funds.” Working Paper, University of Florida.

Kanniainen, V., and C. Keuschnigg, 2000. “The Optimal Portfolio of Start-up Firms in Venture Capital
       finance.” CESifo Working Paper No.381, Journal of Corporate Finance, forthcoming.

Kanniainen, V., and C. Keuschnigg (2001). “Start-up Investment with Scarce Venture Capital Support.”
       CESifo Working Paper No. 439. Journal of Banking and Finance, forthcoming.

Kerins, F., J.K. Smith, and R.L. Smith (2003). “Opportunity Cost of Capital for Venture Capital Investors
        and Entrepreneurs.” Journal of Financial and Quantitative Analysis, forthcoming.

Keuschnigg, C. (2002). “Taxation of a Venture Capitalist with a Portfolio of Firms.” University of St.
       Gallen Working Paper. Oxford Economic Papers, forthcoming.

Keuschnigg, C., and S.B. Nielsen (2001). “Public Policy for Venture Capital.” International Tax and Public
       Finance 8, 557-572.

Keuschnigg, C., and S.B. Nielsen (2003a). “Tax Policy, Venture Capital and Entrepreneurship.” Journal of
       Public Economics, 87, 175-203.

Keuschnigg, C., and S.B. Nielsen (2003b). “Start-ups, Venture Capitalists and the Capital Gains Tax.”
       Journal of Public Economics, forthcoming.

Khorana, A. (2001). “Performance Changes Following Top Management Turnover: Evidence from Open-
       End Mutual Funds.” Journal of Financial and Quantative Analysis 36, 371-393.
                                                                                                      28

Lerner, J. (1999). “The Government as Venture Capitalist: The Long-run Effects of the SBIR Program.”
        Journal of Business 72, 285-318.

Lerner, J. (2002). “When Bureaucrats Meet Entrepreneurs: The Design of Effective ‘Public Venture Capital’
        Programmes.” Economic Journal 112, F73-F84.

Lin, T.H., and R.L. Smith (1997). “Insider Reputation and Selling Decisions: The Unwinding of Venture
        Capital Investments During Equity IPOs.” Journal of Corporate Finance 4, 241-263.

Martin, J.D., and W. Petty (1983). “An Analysis of the Performance of Publicly Traded Venture Capital
        Companies.” Journal of Financial and Quantitative Analysis 18, 401-410.

Massa, M. (2003). “How do Family Strategies Affect Fund Performance? When Performance-
       Maximization is not the Only Game in Town.” Journal of Financial Economics 67, 249-304.

Nanda, V.K., and R. Singh (1998) “Mutual Fund Structures and the Pricing of Liquidity.” Working
       Paper, University of Michigan and University of Minnesota.

Neus, W., and U. Walz (2002). “Exit Timing of Venture Capitalists in the Course of an Initial Public
       Offering.” Working Paper, University of Frankfurt, Center for Financial Studies.

Osborne, D., and D. Sandler (1998). “A Tax Expenditure Analysis of Labour-Sponsored Venture Capital
       Corporations.” Canadian Tax Journal 46, 499-574.

Peng, L. (2001). “Building A Venture Capital Index.” Yale Center for International Finance Working Paper.

Pastor, L., and R.F. Stambaugh (2002). “Investing in Equity Mutual Funds.” Journal of Financial
        Economics 63, 351-380.

Rea, J.D., and B.K. Reid (1998). “Trends in the Ownership Cost of Equity Mutual Funds.” Investment
        Company Institute Perspective 4 (3), 1-16.

Ruckman, K. (2003). “Expense Ratios of North American Mutual Funds.” Canadian Journal of Economics
      36, 192-223.

Sahlman, W.A. (1990). “The Structure and Governance of Venture Capital Organizations.” Journal of
       Financial Economics 27, 473-524.

Schwienbacher, A. (2001). “Innovation and Venture Capital Exits,” Working Paper, University of
       Amsterdam.

Smith, J.K., and R.L. Smith (2000). Entrepreneurial Finance. New York: John Wiley & Sons, Inc.

Wermers, R. (2000). “Mutual Fund Performance: An Empirical Decomposition into Stock-Picking
      Talent, Style, Transaction Costs and Expenses.” Journal of Finance 54, 1655-1703.

Vaillancourt, F. (1997). “Labour-Sponsored Venture Capital Funds in Canada: Institutional Aspects, Tax
        Expenditure and Employment Creation”. In P. Halpern, ed., Financing Innovative Enterprise in
        Canada, University of Calgary Press, 571-592.
                             Figure 1. Capital Under Management by Investor Type in Canada: 1992-2001



                        25




                        20




                        15


Billions of 2001 $Can


                        10




                        5




                        0
                             92       93       94          95           96              97         98           99             00   01
                                                                                 Year


                                               Other Private Equity Fund Types     Labour Sponsored Investment Funds (LSIFs)
                                                                                                                        31

                                  Figure 2. Capital for Investm ent in Canada: 1988-2001



                   25




                   20




                    15

Billions of 2001 Can$

                    10




                        5




                         0
                             92
                                        93
                                                 94
                                                           95
                                                                     96
                                                                                 97
                                                                   Year                   98
                                                                                                    99
                                                                                                              00
                                                                                                                   01

                                  New Capital Raised   Capital for Investm ent   Capital Under M anagem ent
                                                                                                                                                                                             32

                                                                          Figure 3. Selected Indices 1992 - 2002

                          700

                                                                                      The Peng (2001) data stops at 1999. The Venture Economics Post-
                                                                                      Venture Capital Index (PVCI) indicates an index value of 361.36 as at
                          600
                                                                                       06/28/2002 (based on venture-backed companies over the past 10
                                                                                      years). The Peng (2001) index is based on Venture Economics data,
                                                                                      but there are some differences in the index computation methods.
                          500
Total Percentage Return




                          400



                          300



                          200



                          100



                            0
                                                                    95




                                                                                                                            98




                                                                                                                                                                               01
                                      3




                                                3



                                                          94




                                                                                  6




                                                                                                6



                                                                                                              97




                                                                                                                                              9




                                                                                                                                                            9



                                                                                                                                                                      00




                                                                                                                                                                                         2
                                              -9




                                                                                              -9




                                                                                                                                                          -9
                                   -9




                                                                               -9




                                                                                                                                           -9




                                                                                                                                                                                      -0
                                                        p-




                                                                  n-




                                                                                                            p-




                                                                                                                          n-




                                                                                                                                                                    p-




                                                                                                                                                                             n-
                                 ar




                                           ec




                                                                             ar




                                                                                           ec




                                                                                                                                         ar




                                                                                                                                                       ec




                                                                                                                                                                                    ar
                                                                Ju




                                                                                                                        Ju




                                                                                                                                                                           Ju
                                                     Se




                                                                                                         Se




                                                                                                                                                                 Se
                                 M




                                                                            M




                                                                                                                                       M




                                                                                                                                                                                    M
                                          D




                                                                                          D




                                                                                                                                                      D
                          -100
                                                                                                           Date

                                             Globe LSVCC Peer Index                          Globe Canadian Small Cap Peer Index                        TSE 300 Composite Index
                                             US VC Index (Peng, 2001, Figure 7)              30 Day T-Bill Index
                                              Table 1. Summary Statistics: Returns by Characteristics of Labour-Sponsored Investment Funds (LSIFs)


                                                                                                                  Average Returns (annualized) to period ending 1 June 2002
                                                    Number      Average Left Hand        Average 1-         Average 3-      Average 6-      Average 1-     Average 3-    Average 5-          Average 10-
                                                   of Funds       Column Values         Month Return       Month Return      Month Return       Year Return     Year Return   Year Return    Year Return


                 LSIF Average                          50                                   -1.530             -3.000             -4.480           -10.410         1.830          1.950          2.890
     Nesbitt Burns Cdn Small-Cap Index                                                      -0.050             8.910             27.500                9.560      10.400          5.730          10.510


             -0.2 < 3-Year Beta < 0                    3                                    -1.070             -1.637             -3.173           -4.810         -1.740          0.445            ---
              0 < 3-Year Beta < .2                     2                                    -0.735             -1.480             -0.835           -2.585         -0.855          1.650            ---
             .2 < 3-Year Beta < .4                     4          Average Beta of           -0.753             -2.435             -2.085           -9.348          4.600          8.137          7.110
             .4 < 3-Year Beta < .6                     5            All LSIFs =             -3.386             -4.236             -5.284           -18.006         1.814          1.966            ---
             .6 < 3-Year Beta < .8                     1               0.378                -3.490             -5.320             -3.010           -23.780        -10.840        -5.130          -1.340
              .8 < 3-Year Beta < 1                     1                                    -7.800            -12.630            -11.720           -11.760        13.060          6.800            ---
             1 < 3-Year Beta < 1.2                     1                                    -2.120             -7.800            -17.790           -36.560        -15.000        -7.600            ---
          3-Year Beta Not Available1                   33                                   -1.363             -2.745             -3.781           -7.071           ---            ---             ---


         VC Fund Early Stage Focus                     18                                   -0.847             -1.951             -2.077           -7.316          5.820          4.033            ---
    VC Fund General Technology Focus                   8                                    -1.797             -3.617             -4.913           -11.720         7.257          5.020            ---
    VC Fund Specific Technology Focus                  15                                   -1.089             -2.824             -3.888           -6.739          4.430          2.000            ---
   VC Funds part of VC Firm with >1 Fund               38                                   -1.956             -3.376             -5.038           -11.809         0.316          1.171          2.885


                 Age < 1 Years                         15                                   -0.857             -1.039               ---                 ---         ---            ---             ---
            1 Year < Age < 2 Years                     6                                    -1.322             -4.603             -3.842           -7.872           ---            ---             ---
           2 Years < Age < 3 Years                     10                                   -2.194             -3.166             -3.745               -6.591       ---            ---             ---
           3 Years < Age < 4 Years                     1          Average Age of            0.500              0.400              -0.570           -3.240          7.270           ---             ---
           4 Years < Age < 5 Years                     1            All LSIFs =             -0.300             -0.830             0.950                -1.450     -0.720           ---             ---
           5 Years < Age < 6 Years                     4               44.180               -2.277             -1.640             -3.503           -17.813         6.040          3.920            ---
           6 Years < Age < 7 Years                     1             (Months)               -2.120             -7.800            -17.790           -36.560        -15.000        -7.600            ---
           8 Years < Age < 9 Years                     6                                    -3.560             -5.508             -7.023           -11.495         1.328          1.843            ---
                 Age > 9 Years                         6                                    -1.584             -3.668             -1.682           -11.092        -1.976          3.456          2.885


                    Federal                            14                                   -1.334             -1.998             -3.164           -7.523         -1.213         -0.624          -0.096
                    Ontario                            29                                   -1.582             -2.851             -2.814           -6.403          0.207          0.994            ---
               British Columbia                        2                                    -5.255             -9.560             -5.545           -16.625         7.865          5.105          3.555
                Saskatchewan                           2                                    0.030              -0.290             -0.170           -1.245          3.635           ---             ---
                   Manitoba                            2                                    -0.600             -1.410             -0.810           -2.750         -2.125          1.100            ---
                    Quebec                             1                                    0.000              -7.000             -7.000           -11.400         8.600          8.100            ---
                New Brunswick                          1                                    0.000              0.000              0.000                 ---         ---            ---             ---


             Assets < $20 million                      18        Average Assets of          -1.399             -2.215             -3.486           -9.912          2.590          2.735            ---
      $20 million < Assets < $40 million               9            All LSIFs =             -0.968             -1.561             -1.881           -3.869          7.270           ---             ---
      $40 million < Assets < $60 million               2        2002 $Can 79.798            -0.675             -2.390             -3.235           -10.730        -8.660          4.000            ---
      $60 million < Assets < $80 million               9        (Excluding Quebec)          -1.687             -3.648             -4.341           -9.866          4.600          3.560            ---
     $80 million < Assets < $100 million               1          =$Can 168.202             -0.550             -0.080             -4.040           -7.640          8.460          5.300            ---
             Assets > $100 million                     11       (Including Quebec)          -2.905             -5.623             -6.034           -14.454        -0.704          1.394          2.457


                 Cash2 < 33%                           13        Average Cash of            -1.402             -3.952             -5.331           -9.475         -2.356          1.594          8.100
              33% < Cash < 66%                         11           All LSIFs =             -2.293             -3.104             -3.892           -13.312         1.711          3.021            ---
                  Cash > 66%                           5               0.351                -0.290             -1.276             -1.026           -5.803          8.460          5.300            ---


                 Equity2 < 33%                         10        Average Equity of          -0.450             -1.516             -1.452           -4.113          4.110          3.545            ---
              33% < Equity < 66%                       18           All LSIFs =             -2.194             -4.152             -5.526           -13.630         0.516          2.513          8.100
                 Equity > 66%                          1               0.375                -0.900             -1.990             -2.570           -4.050         -3.530          2.200            ---


                 Bonds2 < 33%                          17        Average Bonds of           -1.751             -2.952             -3.197           -11.739         2.096          3.358            ---
             33% < Bonds < 66%                         9            All LSIFs =             -1.509             -3.052             -6.859           -9.684         -4.277          0.120            ---
                 Bonds > 66%                           3               0.274                -0.517             -2.043             -0.157               0.830        ---            ---             ---


1. Betas are actually "pseudo betas" as LSIF share prices are based on periodic (typically quarterly) valuations, where they have been in existence for at least 3 years. The pseudo-betas are
reported on www.globefund.com. There does not exist a market price per se for the funds (independent of the quarterly valuations), and so actual betas and alphas do not exist.
2. Security allocations not known for 21 of the 50 LSIFs. Other/unknown category not reported. The security allocations are reported on www.globefund.com for typical levels held.
"---" means that returns are not systematically publicly reported, or the fund was recently introduced so the data are not available for the period.
Data sources: www.globefund.com, www.morningstar.ca.
                                                                                                                                                                   Table 1 continues on the following page…
                                                                                                                                                                                    34



                                                                          Table 1 Summary Statistics (Continued).


                                                                                                      Average Returns (annualized) to period ending 1 June 2002
                                              Number     Average Left Hand      Average 1-       Average 3-     Average 6-      Average 1-     Average 3-    Average 5-   Average 10-
                                              of Funds    Column Values        Month Return    Month Return     Month Return   Year Return   Year Return    Year Return   Year Return


                 No Load                         6                                 -1.902          -4.685           -3.443       -12.174         6.933         6.837         7.605
                 Back End                       42                                 -1.587          -2.624           -3.028        -9.746         0.616         0.933        -1.340
                 Front End                       1                                 -0.810          -0.700            ---            ---           ---           ---           ---
                 Optional                        1                                 1.560           -1.150           1.880        -12.320        -13.060        8.900          ---


       1% < Manager Fixed Fee < 2%              14       Average Manager           -1.454          -2.264           -2.928       -17.068         0.815         1.740         8.100
       2% < Manager Fixed Fee < 3%              21          Fixed Fee =            -2.170          -4.172           -3.447        -9.799         3.359         3.310         2.885
       3% < Manager Fixed Fee < 4%              12            2.500%               -0.422          -1.089           -1.273        -5.456        -1.025         2.998          ---
         Manager Fixed Fee > 4%                  3                                 -2.117          -2.597           -5.777       -10.610        -4.260         -0.900         ---


       Manager Carried Interest = 0%             5                                 -0.294          -2.082           -2.158        -5.103         1.663         3.298         8.100
   0% < Manager Carried Interest < 10 %          3       Average Manager           -3.280          -4.370           -6.780       -15.000        -7.550         -3.015       -1.340
   10% < Manager Carried Interest < 20%         14       Carried Interest =        -1.375          -2.841           -3.627        -9.927        -4.350         -2.750         ---
      Manager Carried Interest = 20%            23           16.270%               -1.850          -3.110           -2.272       -10.589         3.081         5.825         7.110
      Manager Carried Interest > 20%             5                                 -0.842          -1.066           -2.360       -10.400         4.430         2.000          ---


       0% < Advisor Fixed Fee < 1%              20        Average Advisor          -1.552          -3.921           -3.257        -9.538         1.797         4.116         4.623
       1% < Advisor Fixed Fee < 2%              25          Fixed Fee =            -1.657          -2.441           -3.234       -10.678        -0.487         0.509          ---
       2% < Advisor Fixed Fee < 3%               5            0.922%               -0.974          -0.148           0.000           ---           ---           ---           ---


       Advisor Carried Interest = 0%            37        Average Advisor          -1.168          -2.733           -2.665        -9.927         1.566         3.294         4.623
    0% < Advisor Carried Interest < 5 %         13       Carried Interest =        -1.162          -3.005           -3.645       -10.542        -4.350         -2.750         ---
       Advisor Carried Interest > 5%             0            1.140%                 ---             ---             ---            ---           ---           ---           ---


          Graduated Fixed Feed?                 23                                 -1.893          -3.327           -3.905       -10.710         0.536         2.047         4.623
  Carried Interest Performance Hurdles?         35                                 -1.621          -3.022           -3.146       -10.935         1.923         3.483         7.110


 1% < Total MER in last financial year < 2%      6                                 -0.828          -2.267           -1.668       -17.590        -1.120         1.485         3.380
              2% < MER < 3%                      6        Average MER of           -1.992          -4.395           -2.752       -11.525         9.070         8.190         7.110
              3% < MER < 4%                     11          All LSIFs =            -1.169          -1.195           -1.413       -10.097        -1.565         0.450          ---
              4% < MER < 5%                     13            4.028%               -1.535          -2.866           -2.702        -7.916        -1.792         3.623          ---
              5% < MER < 6%                      9                                 -2.371          -3.804           -6.528       -11.647         1.532         0.298          ---
              6% < MER < 7%                      3                                 -0.380          -1.227           -2.023        -2.917          ---           ---           ---
              7% < MER < 8%                      2                                 -2.555          -1.800           -1.980       -23.530         3.360         4.370          ---



Data sources: www.globefund.com, www.morningstar.ca.
                                                                                                                                                                                                                                                                 35

                                                                                                             Table 2. Fund Pseudo Betas and Risk
This table presents: Models (1) - (4): OLS estimates of the determinants of the fund betas; and Models (5) - (8): OLS estimates of the determinants of the fund risk (return variance) for the period to 6/1/2002. Independent variables:
the percentage of equity and bond holdings held by the fund (a variable for cash holdings is suppressed), the age of the fund (in months), a dummy for funds in firms with more than 1 fund, managerial fixed fee and carried interest %,
dummies for jurisdiction of incorporation (an Ontario dummy is suppressed), dummies for early stage focus, general tech focus, and specific tech focus, dummies fo back end and no load sales fee structure, and variables for the level
of minimum Registered Retirement Savings Plan (RRSP) and subsequent RRSP purchase levels to obtain tax breaks in the RRSP. HCCME estimated covariance matrix. *, **, *** Significant at the 10%, 5% and 1%% levels,
respectively.
                                                                                    OLS: 3-Year Pseudo Betas                                                                           OLS: 3-Year Return Variance (Dependent Variable in Percentages)
                                                     Model (1)                    Model (2)                       Model (3)                     Model (4)                     Model (5)                       Model (6)                    Model (7)                     Model (8)
                                           Coefficient t-statistic      Coefficient       t-statistic   Coefficient t-statistic       Coefficient t-statistic       Coefficient       t-statistic   Coefficient t-statistic      Coefficient t-statistic       Coefficient t-statistic


                Constant                      0.065            0.571      -0.299           -1.678         -0.458          -2.884**      0.781           3.590***     17.637           3.245***        1.697             0.270      -4.988           -1.268      30.231           4.909***


            % Equity Holdings                 0.028            1.267        ---               ---         0.004             1.555       0.004           2.716**       0.063             0.668           ---               ---      0.134            1.915*       0.144           2.872**
            % Bond Holdings                    ---               ---      0.004             1.025           ---               ---       -0.001            -0.502        ---               ---         0.030             0.242        ---               ---       -0.141           -1.685


                Fund Age                      0.001            0.062      0.003            2.008*         0.004           2.910**         ---               ---       -0.077           -1.729         0.026             0.486      0.058            2.086*         ---               ---
 Fund part of Firm with more than 1 LSIF       ---               ---        ---               ---         -0.022           -0.237       0.065             0.596         ---               ---           ---               ---      -1.087           -0.368       2.408            0.645


          Manager Fixed Fee %                  ---               ---        ---               ---           ---               ---       -0.201          -2.924***       ---               ---           ---               ---        ---               ---       -5.756          -3.125***
       Manager Carried Interest %              ---               ---      0.026           4.027***        0.018           2.911**         ---               ---         ---               ---         0.867           6.396***     0.765           4.600***        ---               ---


              Federal Fund                    0.419          2.717**        ---               ---         0.514           3.152**         ---               ---      11.288            2.431**          ---               ---     15.757           4.178***        ---               ---
              Quebec Fund                      ---               ---        ---               ---           ---               ---       -0.254            -1.574        ---               ---           ---               ---        ---               ---      -22.865          -4.480***
          British Columbia Fund                ---               ---        ---               ---         -0.075           -0.415         ---               ---         ---               ---           ---               ---      7.928            1.457          ---               ---
           Saskatchewan Fund                   ---               ---        ---               ---           ---               ---       -0.048            -0.488        ---               ---           ---               ---        ---               ---       1.653            0.457
             Manitoba Fund                     ---               ---        ---               ---           ---               ---       -0.363          -3.489***       ---               ---           ---               ---        ---               ---      -14.530          -3.965***


            Early Stage Focus                  ---               ---        ---               ---         -0.469          -3.585***       ---               ---         ---               ---           ---               ---     -18.066          -8.921***       ---               ---
       General Technology Focus               0.398          3.887***       ---               ---         0.268           3.226**         ---               ---       9.961            2.180**          ---               ---      4.601            1.377          ---               ---
           Specific Tech Focus                 ---               ---      -0.282           -1.834*          ---               ---         ---               ---         ---               ---         -9.958          -2.615**       ---               ---         ---               ---



         Number of Observations                         18                           18                              18                            18                            18                              18                           18                            18
                F Statistic                           3.67**                        1.69                           3.63**                         2.77*                         1.99                           3.28**                       5.37***                       4.20***
              Loglikelihood                           3.023                        -0.006                          9.200                         5.920                        -63.492                         -61.548                      -52.011                       -55.442
       Akaike Information Statistic                   0.220                        0.556                           -0.022                        0.231                         7.610                           7.389                        6.779                         7.049
              Adjusted R2                             0.386                        0.140                           0.553                         0.421                         0.189                           0.349                        0.673                         0.746
                                                                                                                                                                                                                                                                                                                               36

                                                                                                                                                    T a b le 3 . F u n d R e tu rn s

T h is t a b le p r e s e n t s O L S e s t im a t e s o f t h e d e t e r m in a n t s o f t h e c r o s s - s e c t io n a l r e t u r n s a c r o s s f u n d s t o t h e p e r io d e n d in g 6 / 1 / 2 0 0 2 f o r 1 m o n t h , 3 m o n t h s , 6 m o n t h s , 1 y e a r , 3 y e a r s , a n d 5 y e a r s .
I n d e p e n d e n t v a r ia b le s : t h e p e r c e n t a g e o f e q u it y a n d b o n d h o ld in g s h e ld b y t h e f u n d ( a v a r ia b le f o r c a s h h o ld in g s is s u p p r e s s e d ) , t h e f u n d b e t a s , t h e a g e o f t h e f u n d ( in m o n t h s ) , a d u m m y f o r f u n d s in
f ir m s w it h m o r e t h a n 1 f u n d , d u m m ie s f o b a c k e n d a n d n o lo a d s a le s f e e s t r u c t u r e , a n d v a r ia b le s f o r t h e le v e l o f m in im u m R e g is t e r e d R e t ir e m e n t S a v in g s P la n ( R R S P ) a n d s u b s e q u e n t R R S P
p u r c h a s e le v e ls t o o b t a in t a x b r e a k s in t h e R R S P , v a r ia b le s f o r m a n a g e r ia l a n d a d v is o r f ix e d f e e s a n d c a r r ie d in t e r e s t p e r c e n t a g e s , d u m m ie s f o r ju r is d ic t io n o f in c o r p o r a t io n ( a n O n t a r io d u m m y is
s u p p r e s s e d ) , d u m m ie s f o r e a r ly s t a g e f o c u s , g e n e r a l t e c h f o c u s , a n d s p e c if ic t e c h f o c u s . H C C M E e s t im a t e d c o v a r ia n c e m a t r ix . * , * * , * * * S ig n if ic a n t a t t h e 1 0 % , 5 % a n d 1 % % le v e ls , r e s p e c t iv e ly .

                                                                                                       L o g (1 + 1 m o n th re tu rn )                                                           L o g (1 + 3 m o n th re tu rn )                                                           L o g (1 + 6 m o n th re tu rn )
                                                                                              M o d e l (1 )                              M o d e l (2 )                                 M o d e l (3 )                              M o d e l (4 )                                M o d e l (5 )                              M o d e l (6 )
                                                                                 C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t     t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t        t - s t a t is t ic


                                C o n s ta n t                                       0 .0 2 4               1 .2 2 5              0 .0 6 1            3 .0 5 1 * * *            0 .0 2 8               1 .2 1 9              0 .1 1 8            6 .9 2 1 * * *           0 .0 6 5           3 .5 0 0 ***              0 .1 3 1                3 .5 3 3 * * *


                         % E q u it y H o ld in g s                               -0 .0 0 0 0 1             -0 .1 0 0            -0 .0 0 1            -3 .3 2 1 ***           -0 .0 0 0 1              -0 .3 5 9            -0 .0 0 1           -3 .2 4 5 * * *          -0 .0 0 1           -2 .3 9 5 **             -0 .0 0 1               -4 .1 5 9 ***
                         % B o n d H o ld in g s                                        ---                    ---               0 .0 0 0 3            2 .0 1 7 * *                ---                    ---                0 .0 0 1            3 .7 0 6 * * *              ---                    ---               0 .0 0 0 1                 0 .4 9 6


                                    B e ta                                          -0 .0 4 2             -2 .4 5 0 * *          -0 .0 4 1             -1 .9 0 5 *             -0 .0 8 8           -4 .1 6 7 * * *          -0 .1 2 7           -6 .2 7 6 * * *          -0 .1 1 9           -2 .9 9 7 * * *          -0 .1 4 3               -3 .3 6 9 ***
                                Fund Age                                           -0 .0 0 0 4            -1 .9 5 4 *            0 .0 0 0 1              1 .0 1 9             -0 .0 0 0 3          -3 .3 4 7 * * *          0 .0 0 0 1              1 .0 3 7            -0 .0 0 0 3          -2 .3 8 0 **             0 .0 0 0 1                 0 .8 1 6
       F u n d p a r t o f F ir m w it h m o r e t h a n 1 L S I F                      ---                    ---                0 .0 2 3            2 .6 9 1 * * *               ---                    ---                0 .0 3 7            3 .4 2 2 * * *              ---                    ---                0 .0 2 0                  1 .1 0 4


                           B ack E nd Fees                                          -0 .0 0 7               -0 .6 4 6            -0 .0 6 2            -4 .2 9 8 ***            -0 .0 0 2               -0 .0 9 9            -0 .1 0 0           -8 .2 8 2 * * *          -0 .0 2 2               -1 .6 0 8            -0 .0 9 3               -4 .3 5 4 ***
                                 N o Load                                               ---                    ---               -0 .0 3 8            -2 .5 5 3 * *                ---                    ---               -0 .0 9 1           -6 .3 9 0 * * *              ---                    ---               -0 .0 6 7                -2 .2 5 2 * *
              M in im u m I n it ia l R R S P P u r c h a s e                    -1 .0 0 E -0 5           -1 .8 9 0 *          9 .2 3 E -0 6             1 .5 2 3           -3 .9 7 E -0 4             -0 .4 7 9          2 .2 0 E -0 5          4 .4 3 6 * * *       -1 .6 9 E -0 5          -1 .9 2 6 *           5 .9 9 E -0 6                0 .6 0 8
        M in im u m S u b s e q u e n t R R S P P u r c h a s e                         ---                    ---             3 .3 1 E -0 5             1 .7 9 1 *                ---                    ---             5 .5 2 E -0 5          3 .0 5 9 * * *              ---                    ---             6 .3 2 E -0 5               1 .8 0 4 *


                       M a n a g e r F ix e d F e e s                               -0 .0 0 1               -0 .3 0 6            -0 .0 1 0            -2 .1 1 9 * *            -0 .0 0 1               -0 .2 5 1            -0 .0 2 2           -6 .4 1 0 * * *          -0 .0 0 3               -0 .7 7 6            -0 .0 1 8                -2 .2 4 8 * *
                        A d v is o r F ix e d F e e s                                   ---                    ---               -0 .0 0 1               -0 .5 2 6                 ---                    ---                0 .0 2 9           1 1 .2 5 8 * * *             ---                    ---                0 .0 1 9                4 .3 5 8 * * *
                   M a n a g e r C a r r ie d I n t e r e s t                       -0 .0 0 1               -0 .9 5 9            -0 .0 0 1            -2 .4 6 3 * *            -0 .0 0 1               -1 .2 6 2            -0 .0 0 3           -4 .9 2 9 * * *           0 .0 0 0               -0 .0 1 3            -0 .0 0 2                -1 .8 5 7 *
                    A d v is o r C a r r ie d I n t e r e s t                           ---                    ---                0 .0 0 1               0 .6 8 9                  ---                    ---               -0 .0 0 7           -5 .8 7 6 * * *              ---                    ---                0 .0 0 0                 -0 .1 2 5


                             F e d e ra l F u n d                                    0 .0 1 2               1 .7 3 5 *            0 .0 0 1               0 .1 5 1               0 .0 2 3            2 .7 6 4 * * *           0 .0 1 9            3 .1 0 1 * * *           0 .0 1 2                1 .2 4 4         -3 .3 3 E -0 6              -0 .0 0 0 2
                            Q uebec Fund                                                ---                    ---                0 .0 0 4               0 .2 8 0                  ---                    ---               -0 .0 2 3               -1 .3 4 7                ---                    ---               -0 .0 2 9                 -1 .1 5 9
                      B r it is h C o lu m b ia F u n d                                 ---                    ---               -0 .0 4 8            -3 .6 8 3 ***                ---                    ---               -0 .0 2 0            -1 .7 7 4 *                 ---                    ---               -0 .0 1 6                 -0 .6 3 0
                       S a s k a tc h e w a n F u n d                                   ---                    ---                0 .0 6 5            4 .6 2 0 * * *               ---                    ---                0 .1 3 7            9 .9 1 1 * * *              ---                    ---                0 .1 0 7                5 .5 7 5 * * *
                           M a n it o b a F u n d                                       ---                    ---                0 .0 3 4            3 .2 0 2 * * *               ---                    ---                0 .0 3 8            3 .0 9 7 * * *              ---                    ---                0 .0 2 7                  1 .4 5 3
                      N e w B r u n s w ic k F u n d                                    ---                    ---                0 .0 4 0               1 .7 1 7 *                ---                    ---                0 .1 4 6            7 .6 2 9 * * *              ---                    ---                0 .1 0 3                2 .4 4 6 * *


                         E a r ly S t a g e F o c u s                                   ---                    ---                0 .0 1 0               1 .5 1 4                  ---                    ---                0 .0 0 2               0 .5 3 3                 ---                    ---                0 .0 1 0                  0 .9 4 3
                  G e n e r a l T e c h n o lo g y F o c u s                         0 .0 1 2               1 .7 6 3 *            0 .0 2 6            4 .4 9 6 * * *            0 .0 1 7              1 .8 6 6 *             0 .0 5 3            9 .0 1 6 * * *           0 .0 3 0           3 .0 1 8 ***              0 .0 5 6                4 .3 9 4 * *
                  S p e c if ic T e c h n o lo g y F o c u s                            ---                    ---                0 .0 1 4               1 .2 8 1                  ---                    ---                0 .0 2 7            3 .2 8 1 * * *              ---                    ---                0 .0 2 4                  1 .1 9 6



                    N u m b e r o f O b s e r v a t io n s                                         50                                           50                                            50                                           50                                           50                                           50
                                F S t a t is t ic                                                1 .3 7                                      2 .6 9 * * *                                  3 .0 2 **                                  1 2 .8 9 * * *                                 5 .3 1 ***                                   3 .6 5 ***
                             L o g lik e lih o o d                                             1 2 9 .3 0 5                                1 5 1 .6 1 0                                   1 1 1 .9 4 1                                1 6 0 .0 3 8                                  1 0 7 .9 5 1                                1 2 2 .8 0 0
                 A k a ik e I n f o r m a t io n S t a t is t ic                                -4 .7 7 2                                    -5 .1 4 4                                     -4 .0 7 8                                    -5 .4 8 2                                    -3 .9 1 8                                    -3 .9 9 2
                             A d ju s t e d R 2                                                 0 .0 6 3                                      0 .4 3 1                                     0 .2 7 1                                      0 .8 4 2                                    0 .4 4 2                                      0 .5 4 3


                                                                                                                                                                                                                                                                                     T a b le 3 c o n t in u e s o n t h e f o llo w in g p a g e …
                                                                                                                                                                                                                                                                                                             37

                                                                                                                                         T a b le 3 . (C o n tin u e d )
                                                                                     L o g (1 + 1 y e a r re tu rn )                                                            L o g (1 + 3 y e a r re tu rn )                                                           L o g (1 + 5 y e a r re tu rn )
                                                                           M o d e l (7 )                              M o d e l (8 )                                 M o d e l (9 )                             M o d e l (1 0 )                             M o d e l (1 1 )                             M o d e l (1 2 )
                                                              C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t     t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic   C o e f f ic ie n t    t - s t a t is t ic


                         C o n s ta n t                          -0 .0 7 8               -0 .7 8 0             0 .0 2 6                  0 .1 7 3           -0 .0 5 3               -0 .6 0 2             0 .0 4 0                0 .3 1 0             0 .0 3 7                0 .6 5 7             0 .0 3 3                0 .7 5 7


                % E q u it y H o ld in g s                       0 .0 0 0 3                0 .7 1 3           -0 .0 0 1                  -1 .4 4 9         -0 .0 0 0 4              -0 .6 8 3                ---                    ---              -0 .0 0 0 2              -0 .5 4 2                ---                    ---
                 % B o n d H o ld in g s                             ---                     ---               0 .0 0 1                  1 .4 4 3               ---                    ---               -0 .0 0 1               -1 .7 3 4                ---                    ---               0 .0 0 0 2               0 .5 0 0


                             B e ta                              -0 .2 7 9           -3 .6 4 9 ** *           -0 .2 3 8            -2 .6 1 3 * * *          -0 .1 6 4             -2 .2 8 6 **           -0 .1 3 0              -2 .6 6 5 * *         -0 .0 5 9               -0 .9 9 7            -0 .0 9 5             -2 .2 8 9 **
                        Fund Age                                -0 .0 0 0 6              -1 .9 8 1 *          -0 .0 0 1                  -1 .5 7 4          0 .0 0 0 9               1 .5 9 0           0 .0 0 0 0 4              0 .0 4 7                ---                    ---               0 .0 0 0 3               0 .8 6 2
F u n d p a r t o f F ir m w it h m o r e t h a n 1 L S I F          ---                     ---              -0 .0 1 4                  -0 .2 9 3              ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---


                   B ack E nd Fees                                0 .0 3 8                 0 .8 1 5            0 .0 2 2                  0 .2 9 7               ---                    ---                0 .0 1 3                0 .2 0 6                ---                    ---               -0 .0 3 3               -1 .2 7 6
                         N o Load                                    ---                     ---               0 .1 1 7                  0 .5 3 4               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---
     M in im u m I n it ia l R R S P P u r c h a s e          -7 .9 6 E -0 5          -2 .5 2 1 **          5 .0 2 E -0 8                0 .0 0 1               ---                    ---            -8 .9 8 E -0 5             -1 .5 0 7                ---                    ---            -5 .4 9 E -0 5             -1 .2 7 6
M in im u m S u b s e q u e n t R R S P P u r c h a s e              ---                     ---            2 .7 3 E -0 4                1 .5 7 9               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---


              M a n a g e r F ix e d F e e s                      0 .0 4 0                 1 .6 8 0           -0 .0 0 7                  -0 .2 7 1          -0 .0 1 0               -0 .5 3 0                ---                    ---               -0 .0 0 2               -0 .0 9 2                ---                    ---
                A d v is o r F ix e d F e e s                        ---                     ---               0 .0 1 2                  0 .5 8 4               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---
           M a n a g e r C a r r ie d I n t e r e s t             0 .0 0 0                 0 .3 0 6           -0 .0 0 3                  -0 .6 5 3           0 .0 0 4               2 .0 2 2 *            0 .0 0 2                0 .5 3 3             0 .0 0 3             2 .4 1 4 * *            0 .0 0 4             3 .1 3 9 * *
            A d v is o r C a r r ie d I n t e r e s t                ---                     ---              -0 .0 2 0             -1 .9 6 0 * *               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---


                     F e d e ra l F u n d                         0 .0 0 4                 0 .1 3 3            0 .0 0 2                  0 .0 5 2               ---                    ---                   ---                    ---               -0 .0 5 9               -1 .7 9 6                ---                    ---
                    Q uebec Fund                                     ---                     ---               0 .0 3 7                  0 .2 7 3               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---
             B r it is h C o lu m b ia F u n d                       ---                     ---              -0 .0 9 5                  -0 .4 6 3              ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---
               S a s k a tc h e w a n F u n d                        ---                     ---               0 .1 2 7                1 .8 9 4 *               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---
                   M a n ito b a F u n d                             ---                     ---               0 .0 2 8                  0 .3 3 8               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---


                E a r ly S t a g e F o c u s                         ---                     ---               0 .0 0 5                  0 .1 4 6               ---                    ---                0 .1 0 7                1 .4 8 4                ---                    ---                   ---                    ---
         G e n e r a l T e c h n o lo g y F o c u s               0 .0 5 3                 1 .4 4 7            0 .0 9 3                1 .8 5 4 *            0 .1 2 0             2 .9 4 8 * *            0 .1 4 6              4 .4 9 6 ***           0 .0 1 5                0 .5 3 7                ---                    ---
         S p e c if ic T e c h n o lo g y F o c u s                  ---                     ---               0 .1 1 7             3 .2 0 8 * **               ---                    ---                   ---                    ---                   ---                    ---                   ---                    ---



           N u m b e r o f O b s e r v a t io n s                               34                                            34                                           18                                           18                                           16                                           16
                        F - S t a t is t ic                                  4 .7 5 * **                                  3 .0 1 * * *                                   1 .1 2                                        1 .0 1                                      1 .7 9                                       2 .3 3
                     L o g lik e lih o o d                                   4 8 .0 7 8                                   6 1 .8 9 6                                    2 3 .5 8 8                                   2 5 .0 4 6                                   3 2 .1 6 8                                   3 3 .3 8 0
         A k a ik e I n f o r m a t io n S t a t is t ic                     -2 .2 4 0                                    -2 .3 4 7                                     -1 .8 4 3                                    -1 .7 8 3                                    -3 .1 4 6                                    -3 .2 9 7
                                        2
                     A d ju s t e d R                                         0 .5 0 6                                     0 .5 6 2                                      0 .0 4 1                                    0 .0 0 3                                     0 .2 4 1                                      0 .3 4 8
                                                                                                                                                                                                                                                                                   38


                                                                                                                                Table 4. Asset Accumulation
This table presents OLS estimates of the determinants of the cross-sectional asset levels (funds raised) across funds to the period ending 6/1/2002 for funds that have been in existence for more than 1 year (models 1 - 3), 3 years (models 4 - 6), and 5 yeas t
(models 7 - 9). The dependent variable (total assets) is measured in millions of 2001 Canadian dollars. Models 3, 6 and 9 present the results without the Quebec fund, as that fund is an outlier with its large asset base (note that the coefficient estima7 - 9).
unaffected, but the model diagnostics change). Independent variables: the age of the fund (in months), a dummy for funds in firms with more than 1 fund, dummies fo back end and no load sales fee structure, and variables for the level of minimum
Registered Retirement Savings Plan (RRSP) and subsequent RRSP purchase levels to obtain tax breaks in the RRSP, variables for managerial and advisor fixed fees and carried interest percentages, dummies for jurisdiction of incorporation (an Ontario
dummy is suppressed), dummies for early stage focus, general tech focus, and specific tech focus, and variables for the fund's return in the prior yea, 3 years and 5 years. HCCME estimated covariance matrix. *, **, *** Significant at the 10%, 5% and 1%%
levels, respectively.
                                                                        Assets (funds >1 year)                                                                 Assets (funds >3 years)                                                               Assets (funds > 5 years)
                                                     Model (1)                Model (2)             Model (3),Excl.Quebec                  Model (4)                     Model (5)         Model (6),Excl.Quebec                 Model (7)                      Model (8)          Model (9),Excl.Quebec
                                           Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic Coefficient t-statistic


                Constant                    -228.535         -1.787*     -80.778         -0.521      -80.778         -0.521       -298.250         -3.373***    -135.237         -1.480    -135.237           -1.480    -363.459         -3.466***    -266.109          -1.805     -266.109          -1.805


                Fund Age                      2.211          4.192***     2.565         5.118***      2.565         5.118***       3.687           4.222***       2.184          2.594**     2.184            2.594**    4.271           4.065***       3.349           2.241**      3.349           2.241**
Fund part of Firm with more than 1 LSIF        ---               ---     109.777          1.405      109.777           1.405        ---                ---       90.438           1.268     90.438            1.268        ---               ---       146.122          3.390***    146.122         3.390***


             Back End Fees                   -12.661         -0.213     -159.059        -2.005**    -159.059        -2.005**        ---                ---       -2.235          -0.040      -2.235           -0.040       ---               ---       -28.918          -0.592      -28.918          -0.592
                No Load                        ---               ---     -79.016         -0.548      -79.016         -0.548          ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---
    Minimum Initial RRSP Purchase            -0.054          -1.174       -0.087         -1.618       -0.087         -1.618          ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---
 Minimum Subsequent RRSP Purchase              ---               ---      -0.675        -3.579***     -0.675        -3.579***        ---               ---         ---               ---       ---              ---        ---               ---          ---               ---        ---             ---


          Manager Fixed Fees                 45.130           1.437      35.898           1.069      35.898            1.069         ---               ---         ---               ---       ---              ---        ---               ---          ---               ---        ---             ---
           Advisor Fixed Fees                  ---               ---     -43.748         -1.065      -43.748         -1.065          ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---
        Manager Carried Interest              6.698          2.517**      7.528         2.020**       7.528         2.020**        5.692           2.029**         ---               ---      ---               ---      5.201            1.804*          ---               ---       ---              ---
         Advisor Carried Interest              ---               ---     -21.446         -1.664      -21.446         -1.664          ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---


              Federal Fund                   37.942          1.849*      55.832           1.426      55.832            1.426      78.673            1.207        65.758           0.952     65.758            0.952      97.027           1.461        47.022            0.411      47.022           0.411
              Quebec Fund                   4241.560      43.041***     4207.817     32.514***         ---              ---      3976.032       44.002***       4086.953      27.050***       ---               ---     3909.114      36.871***       3920.456       17.056***        ---              ---
          British Columbia Fund                ---               ---     -81.950         -0.586      -81.950         -0.586         ---                ---       93.087           0.722     93.087            0.722        ---               ---          ---               ---       ---              ---
          Saskatchewan Fund                    ---               ---     50.896           0.729      50.896            0.729         ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---
             Manitoba Fund                     ---               ---     38.287           0.518      38.287            0.518         ---               ---         ---               ---       ---              ---        ---               ---          ---               ---        ---             ---


           Early Stage Focus                   ---               ---     -81.485        -2.471**     -81.485        -2.471**        ---                ---       10.808           0.091     10.808            0.091        ---               ---       31.824            0.215      31.824           0.215
       General Technology Focus             189.541          1.839*      277.933        3.086***     277.933        3.086***      77.054            0.818          ---               ---       ---              ---      99.828           1.041           ---               ---        ---             ---
       Specific Technology Focus               ---               ---     117.548         1.997*      117.548         1.997*          ---               ---         ---               ---       ---              ---        ---               ---          ---               ---       ---              ---


   Fund's Return Since Fund Inception        -7.105          -2.115**     -8.091        -2.248**      -8.091        -2.248**       5.015            0.687         9.183           0.754      9.183            0.754      6.025            0.687        12.811            0.973      12.811           0.973



        Number of Observations                          34                         34                          33                             18                            18                         17                           16                             16                          15
                F Statistic                          179.38***                132.09***                       3.48**                       222.23***                     129.80***                     1.48                      191.82***                      153.43***                     1.80
              Loglikelihood                          -201.659                 -185.152                       -180.198                      -106.740                      -107.181                    -101.712                    -95.279                        -94.893                      -89.446
       Akaike Information Statistic                   12.451                   12.068                        12.073                         12.638                        12.909                     12.907                       12.785                         12.862                      12.859
              Adjusted R2                             0.980                       0.987                       0.582                         0.987                         0.984                       0.174                       0.987                          0.986                       0.255

				
DOCUMENT INFO