The economics of entrepreneurship 2: What makes a typical entrepreneur? (III) Survival Robert Cressy

Transcription

1 The economics of entrepreneurship 2: What makes a typical entrepreneur? (III) Survival Robert Cressy Professor of SME and Entrepreneurial Finance Cass Business School Professor Robert Cressy, HUBS April Learning objectives To appreciate the facts of small business risk To understand the causes of small business failure To relate these to the characteristics of the entrepreneur, the business, the industry and the economy 2 1

2 General reading Cressy, Robert C. and David Storey, New firms and their bank, National Westminster Bank of Great Britain, 1994 Storey, Understanding the small business sector, Routledge, 1994, chapters 3, 4 and 7 3 Age and entrepreneurship Specific reading Cressy, Robert, Tax, assistance, compliance & the performance of the Smaller Business, a research report to the Federation of Small Businesses, July 2000 Draft copy on web site Peters, Marjolein, Robert Cressy and David Storey, The economic impact of ageing on entrepreneurship and SMEs, report to the Forward Studies Unit of the European Commission, 1999 Draft copy on website Optimism and entrepreneurship De Meza, David and Clive Southey, The winners curse,. 4 2

3 Definition 1/ Survival: concepts In the STE a business is said to survive if its business account remains open with the originating bank This might be thought to be problematic for a number of reasons: Account switches Between branches Across banks Business closures not associated with branch closures The business is sold not necessarily a failure 5 Concepts cont d In fact, a survey of the reasons for account closure was performed in 1991 and showed that: The vast majority of closures were because the business ceased Solvent closures made up 80-93% Insolvent/bankrupt closures made up 7-20% Virtually negligible were: Branch transfers Bank transfers Other 6 3

4 2/ Survival: Overall rates As we saw from the American data small businesses are rather risky Their chances of survival are much lower than large businesses The STE confirmed this fact for the UK quite starkly: Six and a quarter years after start-up Only 19% or 1 in 5 businesses survive Moreover, the failure rate varies little with macro conditions (see chart overleaf) Interestingly, however, only a minority, about 1 in 4 or 27% of closures, are associated with bankruptcy/insolvency But those that do are not costless to the bank Of these bankruptcies a sizeable majority (three quarters) end up in a write-off (loss) to the bank 7 Chart 1: Survival and the macro cycle Source: Cressy, STE data 8 4

5 Exercise 1: Explaining aggregate closure rates Why do you think that overall closures are independent of the state of the macroeconomy whilst bankruptcies vary with it? 9 Exercise 2: 3/ Factors influencing survival With the help of your partner come up with 3 factors that that might enhance the survival of the small business AND 3 factors that are unlikely to help (or hinder) survival survival chances are measured over the first 2.5 years of the business life 10 5

6 Factors influencing startup survival (Answer to Exercise 2) IMPACT Age of owner Team size Work experience in same sector Vocational qualifications: Trade apprenticeship Business purchase Industry factors NO IMPACT Size: Full-time/Part-time employees Gender of the owner Owners with a degree Prior unemployment Previous business ownership Prior income Assets Borrowing 11 Chart 2: Survival and owner age 2.5-year startup survival rates by age of proprietor Fraction surviving Age of proprietor (years) Source: Cressy, STE data 12 6

7 4/ Owner age and survival Chart 2 shows the relationship between survival and owner age for the 1988 sample Each bar shows the fraction of businesses surviving within a given team age group over a 3.5 year period from inception The overall relationship is n-shaped; survival: Youthful teams have a low survival rate as do retirementbased teams Middle aged teams have the highest survival rate 13 Chart 3: Survival and Team size Startup lifespan by team size STE 1988 cohorts Lifespan (median quarters trading) Team size Source: Cressy, STE data 14 7

8 Owner age con t Specifically, note that The survival rate is a mere 30% for 20 year-olds It rises to 70% for 50 year-olds Then declines again to just over 30% for 65 year-olds Note, however, that in the Chart the sample sizes for the ages of 65 and beyond are quite small (not greater than 12); This means we cannot rely on the shape of the curve to tell us the relationship: it may be due to chance factors rather than a genuine age-dependence For other ages otherwise they are statistically respectable In other words, they can reliably considered to represent a relationship between the age and survival 15 Owner age cont d (David: sample sizes not yet shown on the chart!) It is clear from the chart that the optimal age of a team is 50 years We shall investigate the meaning of this statement shortly 16 8

9 5/ Team size and business lifespan Business lifespan is defined as age at death or closure and is measured in quarters from inception in the chart It should be noted that the sample size for a team of 5 individuals is a rather small at n=7 Whilst this figure may be statistically unreliable the others are respectable i.e. significantly different from zero at the 5% level 17 Team size cont d The evidence from the chart suggests an optimal team size of 3 people The big jump is from 2-3 rather than from 3-4 Hence 4 might also be considered optimal The two medians cannot be distinguished statistically (check this!) 18 9

10 Exercise 3: Interpretation Come up with plausible explanations for why age, team size and work experience enhance survival why assets, size and gender do not 19 6/ Interpreting the survival factors (Answer to Exercise 3) Entrepreneurial age Human capital rises with age up to a certain point Experience rises with age; energy decreases There is a point where these two forces maximise survival Employment opportunities decrease with age Wage rates Increase with tenure (correlated with age) and decrease with decaying skills as a wage earner 20 10

11 Optimism Survival cont d Decreases with age; hence chances of unpleasant surprises decrease with age For other studies of the effect of optimism on entrepreneurship see de Meza and Southey(1996) and references therein. Team size Larger teams (up to a certain point) have Greater stability Greater range of skills 21 Survival cont d Work experience in the area Networks already built up and can be tapped into Skills in the job already acquired Useful in evaluating business opportunities Useful in doing certain jobs oneself in self-employment 22 11

12 Survival cont d Business purchase Networks already exist Customers Brand name/goodwill already exists Suppliers Ditto Industry factors Some industries are more risky than others Competition Technology Life cycle of the product/service Etc 23 7/ Factors irrelevant to survival (More answers to Exercise 3) Business size At startup this is determined by the human capital of the entrepreneur (see chart overleaf) Management skills Larger workforce requires more skills Less experienced/capable individuals can only manager a smaller workforce Finance availability Overdraft borrowing required to finance wages Depends on collateral which rises with age (Housing assets are a concave function of age) Hence the dominant factor in the equation is age 24 12

13 Chart 4: Age and startup size 6 Full-time employees (mean Average age of proprietor (years) Source: Cressy, STE data 25 Irrelevant cont d Note from the chart that The shape of the distribution mirrors that of survival curve by age (see chart 2 above) but peaks at slightly younger age of 45 This suggests what is confirmed by a multivariate analysis, namely, that age explains survival and startup size Startup size drops out of the survival equation once age is included 26 13

14 Irrelevant cont d At each age the size of business started by recessionbased entrepreneurs (1991 starts) is higher than boom-based entrepreneurs (1988 starts) Thus the average business started in a recession tends to be larger, measured in employment terms, than one started in a boom The sample size for entrepreneurs at age is unreliably low so this bar should be disregarded 27 Irrelevant cont d Gender of the entrepreneur It is not obvious why gender should play any role in business survival once human capital and industry sector is controlled for Education of the entrepreneur More relevant is experience Distinguish, however, the sophisticated entrepreneur from the typical entrepreneur (see Bhide and later lectures) Education may play a bigger role for the former, especially in the high tech sector 28 14

15 Irrelevant cont d Prior unemployment of the founders Often unemployment status will (unfortunately for the unemployed) reflect lower human capital However, we can distinguish two subsets of the unemployed: Those with redundancy money (reflecting higher HC) Those without 29 8/ Survival and assets: spurious correlation? Chart 5 shows that human capital (age) is correlated with Assets (housing equity) Borrowing (overdraft limits) Survival What is the likely explanation for these relationships? We reserve a full discussion of the relationships until the next topic, but for the moment note that: If human capital (age) drives asset levels then a concave relationship is to age expected: 30 15

16 Chart 5: Equity, ODs and Survival House equity, OD limits and survival s, % year survival rate Housing equity/1000 Odlim/ Owner age 31 Spurious cont d Assets first increase with age as savings are built up from prior employment become invested (inter alia) in property housing equity increases over time Assets then decline with age as retirement approaches and the entrepreneur and his wife close or sell the business and move into a smaller house Housing equity is then consumed in the retirement years Starting a business may also consume equity 32 16

17 Spurious cont d If bank borrowing rules apply, assets limit borrowing In other words, entrepreneurs will want to consume more than their assets allow Therefore, if the constraints bite, borrowing patterns follow asset patterns Both are a function of owner age Capital constraints would seem to exist If age drives survival, then a concave relationship is again expected (see math theory in the Appendix) This may alter the interpretation of the results as we shall see in the next topic 33 9/ Summary and conclusions The human capital of the entrepreneur and her team considerably enhances new business survival Human capital is also associated with a number of other key factors in the business: the value of the assets available to the business at startup the availability of finance to the business at startup the business initial employment size 34 17

18 Summary cont d This may give rise to the idea that more assets will enhance survival by making more credit available from the banks However, some evidence shows that this correlation of assets and survival may be illusory and have no implication of credit constraints The macro economy is important in business survival but mainly serves to alter the balance between solvent/bankrupt and insolvent/non-bankrupt failures. Insolvencies and bankruptcies are higher in a recession Industry factors are significant but play a more subsidiary role in survival 35 Appendix: Math theory of age and survival 36 18