legal conditions and venture capital governance

© Cumming & Johan (2013) Forms of VC Finance

Legal Conditions and Venture Capital Governance

Cumming & Johan (2013, Chapter 13)

1

© Cumming & Johan (2013) Forms of VC Finance

Issues Addressed in this Chapter

Introduction

Evidence

Conclusion

• What drives corporate governance structures in VC finance?

• How do different legal and institutional frameworks in different countries affect international differences in corporate governance structures?

• Three dimensions:– time from fundraising to deal origination – syndication and co-investment– board seats and security choice

2

© Cumming & Johan (2013) Forms of VC Finance

Prior work

Introduction

Evidence

Conclusion

• Law and Finance– La Porta et al. (1997; 1998); Berkowitz et al. (2003)

• Venture Capital meets Law and Finance– VC Contracts:

• Gilson and Schizer (2003); Kapland Stromberg (203,2005); Lerner and Schoar (2005); Cumming and Johan (2006);

– VC Exits:• Black and Gilson (1998); Schleifer /Wolfenzon (2002); Cumming et al. (2006 );

Bienz/Walz (2007)– Size of VC Markets

• Jeng and Wells (2000); Armour and Cumming (2006 )– Public policy

• Keuchnigg/Nielsen (2001, 2003, 2004)

– Datasets not large (200 or so observations) due to private nature of VC data

3

© Cumming & Johan (2013) Forms of VC Finance

New Insights in this Chapter

Introduction

Evidence

Conclusion

• Complementary Elements of Legal Conditions and VC Governance Worldwide:– Screening– Syndication– Co-investment– Board Seats– Functional Definition of Security

• Data– 3838 VC - backed companies– 39 countries, developed and developing, worldwide– Extensive confidential details not previously studied

4

© Cumming & Johan (2013) Forms of VC Finance

Law and Venture Capital Governance

Introduction

Evidence

Conclusion

• Law quality facilitates– Deal screening time– Syndication– Effectiveness of boards

• Law quality mitigates– Co-investment– Securities requiring periodic cash flows pre-exit

• How to measure law quality– Legality (Berkowitz et al.’s weighted average of La Porta et al.’s

indices)– Legal Origin versus Country Dummy Variables

5

© Cumming & Johan (2013) Forms of VC Finance

Data

Introduction

Evidence

Conclusion

• 3838 VC - backed companies• 39 countries, developed and developing, worldwide

• Source: CEPRES, Germany• Partnership among practitioners worldwide to improve data quality for

decision making in private equity

• Extensive confidential details not previously studied

• Little expected sample bias– Good and poor performing investments– Exited and un-exited investments– Matches profile of VC investments across countries and across time

6

© Cumming & Johan (2013) Forms of VC Finance

Time to Deal Origination

Introduction

Evidence

Conclusion

• Time from starting the VC fund until the first-round lead investment

• Not previously studied in any single or multiple country study

• Hazard models

• Legality index from 20 to 21 (e.g. France versus US) lowers the time until lead first investment by 12%

• Legality index from 10 to 11 (e.g. Philippines versus Indonesia) lowers the time until lead first investment by 24%

• Faster time to deal origination in common law countries by approx. 30%

7

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

© Cumming & Johan (2013) Forms of VC Finance

Time to Deal Origination

Introduction

Evidence

Conclusion

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

# Ent Firms

Average Duration (Holding Period in

Days) until First

Investment

Difference Test

(means)

Median Duration (Holding Period in

Days) until First

InvestmentDifference Test

(medians)

Legality Index > 20.5 2052 718.762

-8.479***

469

p <= 1.21e-08***

Legality Index < 20.5 1140 1042.399 660

8

© Cumming & Johan (2013) Forms of VC Finance

 Time to 1st Lead Investment Model (1): All Model (3): All Model (7): Early Model (9): Early

Market and Legal Factors        

Log (1+ MSCI Return at Inv Date) 2.193*** 2.407*** 2.312** 0.494***Log (Committed Capital Overall Market at Inv Date)

-0.538*** -0.648*** -0.454*** -0.047

Log (Legality Index) 2.298**   2.728***  

Log (GNP Per Capita)   0.714***   0.705***

French Legal Origin Dummy   -0.252***   -0.360***

Germany Legal Origin Dummy   -0.080   -0.280**

Scandinavian Legal Origin Dummy   -0.702**   -0.611**

VC Fund Characteristics        

Log (Fund Manager Experience) -0.044 -0.053 -0.115** -0.188***

Log (Fund Vintage Date) -1.604*** -1.671*** -1.635***  

log (Fund Capital / General Partner) -0.083*   -0.018  

Entrepreneurial Firm Characteristics        

Seed Stage 0.475*** 0.669*** 0.439*** 0.837***

Start-up Stage 0.148 0.282* 0.195 0.282*

Early Stage -0.055 -0.025    

Expansion Stage -0.063 0.081    

Log (Industry Market / Book) 0.428*** 0.342*** 0.403*** 0.535***

Model Diagnostics        

Total Number of Observations 2284 3159 1091 1298

Pseudo R2 0.053 0.060 0.051 0.020

Industry, Year, Country, VC firm dummies includedTable 13.4

9

© Cumming & Johan (2013) Forms of VC Finance

Syndication and Co-Investment

Introduction

Evidence

Conclusion

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

• Syndication– Typically ‘Good’– Decision making, value-added, 2 heads better than 1– Legality 20 to 21: increases the probability of syndication by 3.8%– Legality 10 to 11: increases the probability of syndication by 5.5%– Legal origin ambiguous effect

• Co-Investment– Typically ‘Bad’– One VC fund bails out bad investments of prior VC fund from the same VC

organization– Legality 20 to 21: reduces probability of co-investment by 1.8%– Legality 10 to 11: reduces the probability of co-investment by 3.5%– Legal origin ambiguous effect

10

© Cumming & Johan (2013) Forms of VC Finance

Syndication and Co-Investment

Introduction

Evidence

Conclusion

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

# Ent Firms

Proportion of Syndicated Investments

Difference Test

# Ent Firms

Proportion of Co-

InvestmentsDifference

Test

Legality Index > 20.5 1303 0.413

-0.848

1875 0.167

-4.650***

Legality Index < 20.5 848 0.432 930 0.270

11

© Cumming & Johan (2013) Forms of VC Finance

Syndication and Co-Investment Model (13): Syndication

Model (15): Syndication

Model (19):Co-Investment

Model (21):Co-Investment

Constant -2.239*** -1.847*** -0.873* -1.012**

Market and Legal Factors      

Log (1+ MSCI Return at Inv Date) 0.028 -0.007 -0.104** -0.069

Log (Committed Capital Overall Market at Inv Date) 0.118*** 0.124*** 0.085*** 0.081***

Log (Legality Index) 0.577**   -0.369***  

Log (GNP Per Capita)   0.115** -0.093***

French Legal Origin Dummy   0.218*** -0.005

Germany Legal Origin Dummy   0.021 0.105*

Scandinavian Legal Origin Dummy   0.069 -0.056

VC Fund Characteristics      

Log (Fund Manager Experience) -0.129*** -0.125*** 0.028*** 0.028***

Log (Fund Age at First Investment since Fundraising) -0.012** -0.013** -0.016*** -0.016***

Log (Fund Vintage Date) -0.100*** -0.086*** 0.120*** 0.115***

Log (VC Firm Age)      

Entrepreneurial Firm Characteristics      

Seed Stage 0.174*** 0.190***  

Start-up Stage  

Early Stage 0.140*** 0.147*** 0.002 -0.005

Expansion Stage -0.065** -0.062* 0.017 0.008

Log (Industry Market / Book) 0.125*** 0.118*** 0.047*** 0.056***

Investment Characteristics      

Log (Amount Invested) -0.006 -0.005 -0.002 -0.002

Sample Selectivity Correction      

Model Diagnostics        

Total Number of Observations 3176 3159 3176 3159

Pseudo R2 0.212 0.214 0.266 0.271

Industry, Year, Country, VC firm dummies includedTable 13.5

12

© Cumming & Johan (2013) Forms of VC Finance

Board Seats and Functional Security

Introduction

Evidence

Conclusion

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

• Board seats– VC board representation more important than that of other investors

(Gompers and Lerner 1999 MIT Press)– Legality 20 to 21: increases the probability of board seats by 4.2%– Legality 10 to 11: increases the probability of board seats by 8.1%

• Functional security– Security mandates periodic cash flows paid back to the investor prior to

an exit event– Not previously considered in any VC study for single and/or multiple

countries– Legality ambiguous effect– English legal origin approx 16% less likely to require periodic cash flows

13

© Cumming & Johan (2013) Forms of VC Finance

Board Seats and Functional Security

Introduction

Evidence

Conclusion

1. Deal Origination2. Syndication and Co-Investment3. Board Seats and Functional Definition of Security

# Ent Firms

Proportion of

Investments with Board

SeatsDifference

Test# Ent Firms

Proportion of

Investments with

Periodic Cash Flows

Difference Test

Legality Index > 20.5 828 0.722

-0.240

2052 0.335

-5.072***

Legality Index < 20.5 378 0.728 1140 0.429

14

© Cumming & Johan (2013) Forms of VC Finance

Control and Cash Flow Rights Model (25)Board Seats

Model (27)Board Seats

Model (31) Conv-ertible Security

Model (33) Conv-ertible Sectrity

Constant -1.798** -3.797*** -2.121*** -2.595***

Market and Legal Factors        

Log (1+ MSCI Return at Inv Date) 0.096 0.074 0.257*** 0.079

Log (Committed Capital Overall Market at Inv Date) 0.050*** 0.049*** 0.018 0.043**

Log (Legality Index) 0.852***   -0.098  

Log (GNP Per Capita)   0.446***   0.022

French Legal Origin Dummy   0.133***   0.160***

Germany Legal Origin Dummy   -0.058*   -0.023

Scandinavian Legal Origin Dummy   -0.224**   0.188*

VC Fund Characteristics        

Log (Fund Manager Experience) -0.119*** -0.113*** -0.046*** 0.024

Log (Fund Age at First Investment since Fundraising) -0.004 -0.005 -0.016*** -0.029***

Log (Fund Vintage Date) -0.140*** -0.127*** 0.224*** 0.184***

log (Fund Capital / General Partner)       -0.043***

Entrepreneurial Firm Characteristics        

Seed Stage 0.290*** 0.282*** -0.110*** -0.135***

Start-up Stage -0.263*** -0.252*** -0.236*** -0.142***

Early Stage 0.186*** 0.182*** -0.191*** -0.124***

Expansion Stage 0.091*** 0.090*** 0.009 0.024

Log (Industry Market / Book) 0.205*** 0.200*** -0.115*** -0.126***

Investment Characteristics        

Lead Investor 0.296*** 0.280*** 0.125*** 0.093***

Log (Amount Invested) -0.055*** -0.053*** 0.042*** 0.055***

Sample Selectivity Correction        

Model Diagnostics        

Total Number of Observations 3176 3159 3176 2267

Pseudo R2 0.321 0.321 0.207 0.195

Industry, Year, Country, VC firm dummies includedTable 13.6

15

© Cumming & Johan (2013) Forms of VC Finance

Robustness Checks

Introduction

Evidence

Conclusion

• Country dummy variables for English, US, France, German versus legal origin

• Alternative explanatory variables– Market conditions, e.g., MSCI, VC capital in market– VC fund characteristics– Entrepreneurial firm characteristics– Transaction characteristics

• Exclusion of certain countries – e.g., exclude developed countries

• Exclusion of different years – e.g., exclude all years prior to 1995

16

© Cumming & Johan (2013) Forms of VC Finance

Related Evidence

Introduction

Evidence

Conclusion

• Chapter 22

• Better legal standards higher VC/PE returns

• Better legal and accounting standards less likely to observe inflated values of unexited investees reported to institutional investors prior to exit

• Based on data from 39 countries around the world• >5000 transactions

17

© Cumming & Johan (2013) Forms of VC Finance

Concluding Remarks

Introduction

Evidence

Conclusion

• Law quality facilitates– Deal screening time– Syndication– Effectiveness of boards

• Law quality mitigates– Co-investment– Securities requiring periodic cash flows pre-exit (in conjunction with high-tech

investment)

• Legality has a more pronounced effect on emerging markets

• Effective VC investment will continue to lag behind in emerging countries whilst legal quality remains deficient

18