Non Refereed Book Publication 525 Full Text

7/30/2019 Non Refereed Book Publication 525 Full Text

http://slidepdf.com/reader/full/non-refereed-book-publication-525-full-text 1/41

RESEA RCH REPORT 0002 / E

Niels Bosma

Mir jam van Praag

Gerrit de Wit

June 2000

Determinant

s

ofS

ucc

ess

ful E

ntrepreneurs

hip

SCALES SCientif ic AnaLysis of Entrepreneurship and SME s



ISBN: 90-371-0781-8

Price: NLG 35, -

Order nu mber: H0002

This report has been wr it te n in the f rame work of EIM’s economic mod el l ing pro-

gramme SCALES, w hich is f inanced by the Nether lands Ministry of Economic Af fa i rs.

EIM / Bu sin ess & Policy Research employs 170 professionals. EIM provides poli-

cy- and practice-oriented socio-economic information on and for all sectors in

private enterprise and for policy-makers. EIM is established in Zoetermeer.

Besid es on the Netherlands, EIM also focuses on the European economy and

on other continents. You may contact us for more information about EIM and

it s services.

Address: Italiëla an 33

Mail ing address: P.O . Box 7001

2701 AA ZoetermeerTele ph on e: + 31 79 341 36 34

Fa x: + 31 79 341 50 24

Websi te: www.eim.nl

The re s pon s ibil ity for the c ontent s of thi s report l ie s with EIM .

Quot ing o f number s and / or text s a s an explanat ion or s uppor t in paper s, e ss ay s and

book s i s permit ted only when the s our c e i s c lear ly ment ioned .

No part of th i s publ i c at ion may be c opied and / or publ i s hed in any form or by any

mean s, or s tored in a ret r ieval s y s t em , witho ut the pr ior wr it te n permi ss ion of EIM .

EIM do e s not a cc ept re s pon s ibil ity for print ing error s and / or other imperfe c t ion s.



Contents

Summary..........................................................................................

5

1 Introduction .....................................................................................9

2 Theories of entrepreneurship ...................................................11

2.1 Introduc t ion . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .11

2.2 Cl assic v iews on ent repreneurship . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .11

2.3 Three measures of entrepreneur ia l succ es s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

2.4 Determinants o f success . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .14

2.5 Linking determinants t o success . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .14

3 Entrepreneurship data for the Netherlands ..........................16

3.1 Introduc t ion . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .16

3.2 EIM f irm founders survey . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .16

3.3 Entrepreneur ia l success: operat ional var iables . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .17

3.4 Determinants o f success: operat ion al variables . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .18

3.4.1 Human capita l . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . .18

3.4.2 Financial capita l . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. .20

3.4.3 Social capita l . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .21

3.4.4 Keeping up w i th deve lopments re levant for the businesses . . . . . . . . . . . . . . . .22

3.4.5 Controls . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .22

3.5 Discussion . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .23

4 Emp irical results ...........................................................................26

4.1 Introduc t ion . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .26

4.2 Non-response . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . .26

4.3 Profit . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .27

4.4 Employment . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . .29

4.5 Dura t ion . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .31

Literature ........................................................................................34

AppendixThe survival model . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .36



4



5

Summary

I ntrodu c t i on

In the past decades, the v iews on ent repreneurship deve loped towards a

common agreement w i th respec t to i ts impor tance fo r (economic) soci e t y. En-

t repreneurship receives substant ia l interest f rom both sci en t ists and po l icy

makers. Entrepreneur ship , however , has many aspec ts. This diversion can be

seen in li terature - when explor ing th e v iew s of the classic cont r ibutors to en-

trepreneurship t heory - as well as in the real wor ld .

By st imulat ing entrepreneurship , the ac tual goal is especially to st imula te the

people w ho wil l be succ essfu l af ter star t ing up their businesses. Consider ing

the great var iety within the group o f (pot ent ia l) entrepreneurs, there is need

for speci f ic insight in the fac tors that determine the success of star t ing

entrepreneurs. This is invest igate d in the present study for the Nether lands.

Approa c h

We have access to the EIM f irm founde rs survey, a large Dutch data panel con-

sist ing of people who star ted a business in 1994. The f irst survey was held in

the per iod of the start-up . The responden ts gave de ta i led in format ion on

t hemselves, on their environment and on their st rategies. The f irm founders

were fol lowed in the years a f te r, in which they provided informat ion on their

achievements annual ly . Success could be measured by the prof i ts of the en tre-

preneur, employment created by the entrepreneur , and the survival period of

the f irm . The determinants ident i f ied f rom the survey are classi f ied in a f ramework that

dist ingu ishes the dif ferent resources f rom which the star t ing entrepreneur can

draw . These re late to human capita l , f inancia l capita l and socia l capita l . Add i -

t ional ly , t he st ra teg ies of the entrepreneur to keep up with busi ness as well as

some control var iables are ident i f ied .

To invest iga te the in f luences of poten t ia l de terminants (r esult ing f rom the

1994 survey) on success (result ing f rom the surveys until 1997), mult iple regre s-

sion analysis w as used . This approach measures the in f luence of each determi-

nant on succ ess, while control l ing for the other determinant s ident i f ied. For

each succ ess measure , the re lat ions between de terminants and succ ess are es-t imated.

R e s u l t s

Gen eral resul ts are tha t the amount o f human capita l is especia l ly importan t

for determining durat ion and prof i t , while f inanci al capita l is especially re-

la ted to employment . Socia l capita l and st rategies for ret r iev ing re levant in-

format ion seem to be a bou t equal ly important for a ll success measures. The

speci f ic determinants are dealt with below .



6

Human c apita l

Entrepreneurs at higher ag e (age is used as a proxy to ‘knowled ge of the

world’) seem to make less prof i t and to crea te less employment than younger

ent repreneurs do . However, the average durat ion of the founded f irm is

h igher : younger entrepreneurs have higher probabi l i t ies to quit the business

early . The educat ional level of the entreprene ur a lso mat te rs for succ ess,

though only when prof i ts are considered .

Exper ience is also important in determining succ ess. Having had exper ience in

t he same sec tor as the newly founded business increases the probabi l i t ies o f

succ ess in making prof i ts and in surviving. Having exper ience as an employee

has a posi t ive impact on the durat ion o f the f irm, not on prof i ts or employ-

ment . Experience in self -employment only matters for achieving higher prof i ts.

F inancial experience is seen to be less favou rable for the durat ion of the f irm,

whi le i t is posit ively associ a t ed w i th employment .

Finan c ia l c apita l

The amount o f income , other than the income genera ted f rom the founded

firm , has a negat ive e f f ec t on prof i tabi l i ty and generated employment . A f irm

that is f inanced w i th own capita l a lso leads to less employment. Employment

achievements are found to be high whe n a business partner made some f inan-

cial contr ibut ion . No signif icant e f f ec ts were foun d for any of the se determi-

nan ts re lat ing to durat ion .

S o c ia l c apita l

The in f luence of other entrepreneurs in the family is nega t ive ly re lated to

prof i t making . Having con t ac t with other entrepreneurs in ne tworks is posi-

t ive ly re lated to the amount of employment crea ted . Emot iona l support f rom

the spouse inf luences prof i tabi l i ty and durat ion in a posi t ive way. I f the entre-

preneur boards out ac t ivit ies to other part ies, t his can be seen as an indicat ion

o f success regard ing ge nera ted employment .

S t ra teg ie s for keep ing up wi th bu s ine ss

When th e ent repreneur focuses on commercia l re lat ions in ret r iev ing re levant

informat ion th at wi l l he lp to keep up with business, t his indica t es success f or

a l l three measures. Focus on the branch in gen eral is only associ a t ed w i th du-

rat ion , while the focus on direc t business re lat ions (customers and suppl iers) is l inked to prof i tabi l i ty. Informal con t ac t with fe l low-entrepreneurs has a sl ight

in f luence on genera ted employment .

Control variable s

Our invest iga t ions suggest that male entrepreneurs perform bet ter than fe-

male ent repreneurs, though on ly when survival of the f irm is addressed . For

prof i tabi l i ty and employment , no signif icant e f f ec t for gender is f ound . Entre-

preneurs who ind ica ted to have employment growth as a goa l indeed crea ted



7

signif icant ly more employment than the others did . However , they did not

make high er prof i ts. Being part-t ime entre preneur is also important for de-

termining succ ess. Entrepreneurs who are 100 percen t self -employed score es-

pecia l ly higher on d urat ion . Remarkably , a st rong negat ive re lat ion was f ound

in re lat ion to generated em ployment for ful l- time entrepren eurs. An ent r e-preneur who is ac t ive in the business services sec tor, or an entreprene ur who

considers the (expected) higher income as an important mot ivat ion to star t the

busi ness, does not seem to have more succ ess than his counterpar ts.



8



9

1 Introduction

In the pas

t decades,

the v iews

on ent repreneurs

hip deve loped towards

acommon agreement w i th respec t to i ts impor tance fo r (economic) soci e t y. En-

t repreneurship receives substant ia l interest f rom both sci en t ists and po l icy

makers. Extensiv e stud ies are useful as ent repreneurship has many face ts; the

re lated aspec ts are also diverse . This diversion can be seen in l i terature - wh en

explor ing the v iew s of the cl assic cont r ibutors to ent repreneurship th eory - as

well as in the real wor ld .

By st imulat ing entrepreneurship , the ac tual goal is especially to st imulate the

people w ho wil l be succ essfu l af ter star t ing up their businesses. These f irm

founders are bel ieved to play a k ey role in the economic progress. Consider ing

the great variety within th e group of (pote nt ia l) entrepre neurs, th ere is need

for speci f ic insight in the fac tors that determine the success of star t ing

entrepreneurs. This is invest igate d in the presen t study using a longitudinal

pane l da ta se t consist ing of Dutch f irm foun ders.



10



11

2 Theories of entrepreneurship

2.1 Introduction

Al though i t is a lmost unan imously acc epted tha t ent repreneurship plays a ma-

jor role in economic deve lopment, the evidence of this re lat ion is remarkably

l imited . This may very well be caused by the fac t that th is re lat ion requires the

acknowled gement o f in te rmedia te processes. Ent repreneurship is c losely re-

la ted to small businesses and technolog ical change . There is empir ical ev idence

of the posi t ive inf luence of the presence o f small businesses on economic

growth . In order to improve alloca t ion e f f icient ly , potent ia l ly successfu l entre-

preneurs should be st imulated in th eir ac t ivit ies. To be able to dist inguish be-

tween high- and low-potent ia l entrepreneurs, the success determinants of the

entrepreneur wi l l be iden t i f ied in the fol low ing chapters.

2.2 Classic views on entrepreneurship

Economic theor ies at t r ibute a grea t var iety of fun c t ions to the ent repreneur. A

number of func t ion aspec ts appear in many dif ferent theor ies as a part of the

to ta l construc t of the entrepreneur. In order to gain insight in the impor tance

of these recurring fu nc t ion aspec ts, the role played by these aspec ts in the

theor ies of the major cont r ibutors to ent repreneurship research is summarised

in table 1. O f course , this representa t ion abstrac ts f rom many aspects of these

theor ies. This representa t ion is chosen for reasons o f comparabi l i ty and clar ity .

The conceivers of the theor ies summarised in table 1 are widely acc epted as

major cont r ibutors to theore t ica l knowledge about the ent repreneur (see e .g . Hébert and Link , 1982, or van Praag , 1999).

Tab le 1 Aspe cts of the e ntrepreneurial function in theories of the entrepreneur*

period risktaker arbitrageur capital ist manager innovat or

Cantillon 1680-1734 +++** ++ + o o

Say 1767-1832 ++ o - +++ ++

M arshall 1842-1924 + ++ +++ +++ o

Me nger 1840-1921 - + - ++ o

Kni gh t 1885-1972 +++ o - - +

Schump eter 1883-1950 - + - - +++

Kirz ner 1930- -*** +++ - o o

* This summ ary is ba sed o n Cant illon [1931], Say [1845], M arshall [1961], Me nger [1950], Knight[1921], Schum pet er [1934], Schump eter [1942], Kirzn er [1979] and Kirzner [1981].

** The meanin g of the symbols used to summarise the function aspects is:

- : aspect explicit ly excluded

o : aspect not included

+ : aspe ct implicitly included

++ : aspect explicit ly included

+++ : aspect is essential to the theory.



12

*** This denial of the risktaking function by Kirzner refers to the concept s pr esented in Kirzner[1979]. In a lat er pa per (Kirzn er [1981]) risktaking is introduced in his vision on theentre preneurial funct ion.

Table 1 sho ws that r isk tak ing is of ten thought to be included in the entrepre-

neur ia l func t ion . Menger, Schump eter and Kirzner expl icit ly st a te tha t the

f unc t ion of their entrepreneur does not involve tak ing r isks. According to Can-

t i l lon and Knight , ri sk tak ing is the key func t ion o f the en t repreneur. Cant i l lon ,

M arshal l, Menger, Schump eter and Kirzner ident i fy their entrepren eur as an

arbit rageur . Kirzner considers arbit rage ac t ivit ies to be of pr incipa l importance

to the ent repreneur. The quest ion whe ther or not the entre preneur is a capi-

ta l ist is one of th e major issues in entrepreneurship theor ies. The associa t ion

of the entrepreneur with the capita l ist receives st rong support (Marshal l and

Cantil lon) as well as st rong opposi t ion (Say , Menger, Knight , Schumpeter and

Kirzner) . M arshal l even considers the capita l ist func t ion to be an essent ia l as-

pec t of the ac t ivit ies of the en t repreneur. The manager ia l func t ion of the en-

trepreneur is stressed by Say , M arsha l l and Me nger. Knight and Schumpeter

expl ic i t ly deny the associa t ion of the entrepren eur ia l and man ager ia l func t ion .

Both Say’ s and Marshal l ’s entrepre neur are predo minant ly involved in man-

agement ac t ivit ies. There is no opposi t ion against the inclusion of the inn ova-

tor func t ion aspec t in the ac t ivit ies of the ent repreneur . The innovator func-

t ion is, o f course , essent ia l in the entrepreneurship theory of Schumpeter . For

a more deta i led descr ipt ion of the func t ion of the en trepreneur in economic

theor ies see H ébe rt an d Link (1982), Van Dijk a nd Th urik (1995) or van Praa g

(1996, 1999).

2.3 Three measures of entrepreneurial success

The entrepreneur does not exclusive ly determine th e struc ture of a f irm . The

environment a lso plays an important role in determining the struc ture of the

f irm . In academic l i terature a great deal of at tent io n is paid to the inf luence o f

the environmen t on the f irm. Figure 1 sho ws the re la t ion be tw een the ent re-

preneur and success when the impac t of the environmen t on the f irm is ac-

know ledged .



13

Figure 1 The relat ion betw een the entrepreneur and success when th e role of the environmentis acknowledged

Entrepreneur Firm Success

Environment

The entrepreneur and the environment determine the struc ture of the f irm .

Success of the entrepren eur is measured by success of the f irm . The objec t ive

of this study is to ident i fy the success determinants of the ent repreneur . To

correc t for the inf luence of the environmen t a number of var iables characteris-

ing the environment should be included .

P rof i t

The associ a t ion be tween the successfu l entrepreneur and prof i t making is al-

most undispute d in li terature . Table 2 is tak en fro m van Dijk (1996) and sum-

marises the impor tance at t r ibuted to prof i t making in dif ferent entrepreneur-

ship theor ies in a manner simi lar to the summary of func t ion aspec ts of the

entrepreneur in Table 1. Table 2 Importance of prof it making in the t heories of the entrepreneur*.

Cantillon Say Marshall Menger Knight Schumpeter Kirzner

Importance of prof it + +** + ++ ++ +++ ++ +++

* This summ ary is based o n Can tillon (1931), Say (1845), M arshall (1961), Me nger (1950), Knight(1921), Schump ete r (1934), Schump eter (1942), Kirzn er (1979) and Kirzn er (1981).

** The meaning of the symbols used to summarise the importance of profit is:

+ : profi t makin g is associated with the entrepreneur

++ : profit making is considered to be an important result of entrepreneurial act

+++ : profit ma king is t he central issue of the theory.

The emphasis on prof i t making as an indicator of success for the entrepreneur

mot ivates the use of prof i t making a s a performance indicator in this study.

Generated emp l oyment , s urv i va l t i me

Prof i t making re lates to indiv idual success. However , we are a lso interested in

‘success f or socie ty ’. Therefore , prof i t making w il l not be th e only indicator of

entrepreneur ia l success t aken in to considerat ion . This study will also analyse



14

genera ted employment and the survival per iod of the f irm founde d by the en-

trepreneur as a measure of performance .

2.

4 Determinants of success In this sec t ion we explore the determinants of successfu l entrepreneurship

that ar ise f rom l i terature . The cl assic v iews on ent repreneurship produce pos-

sib le success determinants for st a r t ing ent repreneurs. These are summarised in

Table 3, w hich is tak en fro m van Praag (1999)1

. In this study we shall empiri-

cally test proposed de terminants tha t stem from these cl assical v iews, as well

as some other de terminants proposed in l i terature . Some genera l classi f ica t ion

of the de terminants will be clar i fy ing. We make a dist inc t ion be tw een de ter-

minants that are re lated to human capita l , f inancia l capita l , social capita l and

other (control) determinants.

Table 3 Det erminant s of successful entrepreneurship discussed by the classic authors (taken from vanPraag, 1999)

Having success as entrepreneur Start and be a successful entrepreneur

Canti l lon Alertness and foresight

Bear risk

Say Judgement, perseverance , knowledge of

the world, bu sin ess and occupation

Bear risk

M arshall Intelligence, general abili ty (dep endent on

family background and educ ation)

Knowledge of the t rade

Bear Risk

Leadershi p

O w n capi ta l

Good fortune

Father entrepreneur

Schumpeter Leadershi p

Knig ht Abil i ty to deal with uncertainty: self-

confidence , foresight , intel lectual capa city

Good luc k

Kirzner Creat iveness and leadership to exploit

prof it opportunit ies

2.5 Linking determinants to success

To study the impac t of possib le determinants o f succ essfu l entrepreneurship

empirically , the obvious approach is to make use of a panel data set of entre-

preneurs. Determinant var iables are construc ted f rom the charac terist ics of the

ent repreneurs, and success var iables are construc ted f rom th e f irm’s achieve-

1 Van Praag also l isted the determinants f or starting a firm. Th is is not relevant in the present

study, as we w ork with a data panel of entrepreneurs who already started their busin esses. Details of this panel w ill be discussed in section 3.2.



15

ments as reported by the entrepreneur . When the number o f ent repreneurs is

considerable , the analysis wil l lead to useful insights.

We have a sui tab le da taset for the Nether lands at our disposal , w hich is dis-

cussed in chapt er 3. In chapt er 4, the models that are used for l ink ing the de-

te rminants to success are descr ibed .



16

3 Entrepreneurship data for the Netherlands

3.1 Introduction

In the Nether lands, there has been increased a t t ent ion f or ent repreneurship .

This is i l lust rated by publ icat ion o f the government White Paper (1999), de-

scr ib ing the importance o f ent repreneurship for socie ty . The Dutch govern-

ment pol icy is t o st imulate people to star t a f irm and to create an environment

tha t has no unnecessary barriers tha t w ould prevent the en t repreneurs to pe r-

form wel l. To achieve this, i t is of interest to have some knowledge o f the

typical charac terist ics of ent repreneurship in the Nether lands. The const ruc t ion

of the EIM f irm foun ders survey in 1994 has been one of the inst ruments to

achieve this know ledge. A descr ipt ion of this survey is provided in sec t ion 3.2.

From the survey, we are able to const ruc t the var iables as discussed in the pre-

v ious chapt er : measures o f success (sec t ion 3.3) an d possib le de terminants o fsucc ess (sec t ion 3.4).

3.2 EIM firm founders survey

EIM Business & Policy Re search star ted a f irm founders survey in 1994, part ly

on request of the Nether lands Ministry of Economic Af fa i rs. This survey was

const ruc ted to be able to observe posi t ive and negat ive developments in the

Du tch f irm popula t ion . It consists of informat ion of annual interv iew s con-

duc t ed among a sample of star t ing entrepreneurs in the Nether lands. The ini-

t ia l sample was t aken f rom the da tabase of the Dutch Chamber of Commerce . The sample was take n f rom al l new ly registered f irms in the f irst quarter of

1994. The ini t ia l sample consisted of 10,627 firms. Firms included in th is sample

had to sa t isfy a number of speci f ica t ions.

• The f irms had to be foun ded in the f irst quart er of 1994.

• Firms that re-regist e red because of a change in Chamber of Commerce dis-

trict were excluded .

• Firms that re-regist e red because o f some other change in location were ex-

cluded .

• Firms that re-regist e red because of a change in ac t ivit ies or a chang e in le-

gal st a tus were excluded .

• Firms that were parts of previously exist ing companies were excluded .

• Parts of exist ing f irms that regist e red separate ly were excluded .

• Agricultural f irms and f irms in the mineral sec tor were excluded .

To make sure that the f irms were reasonably dist r ibuted amon g the d i f f e rent

sec tors, the f rac t ion of new f irms selec ted f rom the services sec tor was l imited

to one half . Normally, a larger frac t ion of a l l new ly registered f irms be longs to

t his sec tor. This approach caused the o ther sect ors to be represented in the

sample wit h a higher f rac t ion than in the populat ion .



17

The init ial 10,627 firms were con t ac ted by te lephone . A tot a l of 3,000 firm

founders agreed to part icipa te in the survey . Ap proximately 2,000 f irm foun-

ders f ina l ly complet ed the 1994 quest ionnaire . The 2,000 f irm found ers w ho

comple ted the June 1994 quest ionnaire were sent th e 1995 quest ionnaire ,

w hich w as complet ed by over 1,100 responden ts. The f irm si ze and sec tor dis-tr ibuti on o f the 1994 and 1997 responden ts were comparable to those of the

init ia l sample.

The f irst quest ionnaire consisted of 90 quest ions div ided in the categor ie s

genera l, f irm , work exper ience , motives, founding si t ua t ion , capita l and in-

vest men ts, problems encountered , control and subcont rac t ing , env i ronment

and marke t , sales and vulnerabi l i ty , t a rge ts and st ra tegy, performance and ex-

pec t a t ion , and age and educa t ion. The quest ions were based on re levant as-

pects of new venture creat ion as descr ibed in recent l i terature . The quest ion-

naire in 1997 contain ed a tota l of 44 quest ions, in approximately the same ca-

tegor ies. The informat ion f rom the f irst quest ionnaire is used to construc t the entre-

preneur ia l func t ion var iables and the var iables tha t measure the success de-

terminants. Entrepreneur ia l succ ess is exclusively measured by var iables con-

struc t ed f rom the quest ionnaire s in 1995-1997.

3.3 Entrepreneurial success: operational variab les

In chapt er 2 we st a ted our in tended succ ess measures. These three measures

can be crea ted f rom the da taset with informat ion on performance in the pe-

rio d 1994-1997.

P rof i t

Profit is equa ted to th e prof i t made in 1997. The entrepreneur has then been

ac t ive for three years. The profit may especially in the f irst two years be

somewhat misl ead ing , as init ial (sunk) costs of ten have to be ga ined ba ck ,

w hich reduces prof i t . For entrepreneurs who are known to have ended the i r

businesses the prof i t var iable is equated to zero .

E mp l oyment

Where the prof i t measure is mainly an indiv idual succ ess measure , tota l em-

p loyment crea ted can especially be seen as succ ess for socie ty . The employ-ment measure used is t he cumulat ive employment create d in the per iod 1994-

19971

.

1 Other employment measur es have also been invest igated. Th ese i nclude employment grow th, the em ployment in 1997. Both measur es produced similar resul ts. Th is could be expect ed, as high correlations existed between these employment measur es. We therefore chose to focus on the measure that ref lects the cumulat ive employment.



18

Durat i on

Anoth er in te rest ing succ ess measure is t he simple matter of survival. Is the f irm

st i l l in business in 1997? And if n ot , how many months has i t taken before the

entrepren eur quit? In the f irm founders pane l , informat ion is avai lable on the

durat ion of th e f irms. We have construct ed a var iab le mea suring the nu mber

of months that a f irm has been ac t ive . A durat ion mod el wi l l be appl ied to this

variable in chapter 4.

Table 4 Descriptive stat istics of success me asur es

Variable Mean St d . Dev. Min . M ax.

profit in 10 000 guilders 4.3 8.5 0 50

cumulated employment in FTE 3.1 15 .1 0 312

durat ion in months 33 .6 14 .2 0 42

Table 5 Correlations of success measur es

Variable prof it employme nt durat ion

profit in 10 000 guilders 1

cumulated employment in FTE 0.44 1

durat ion in months 0.27 0.11 1

3.4 Determinants of success: operational variab les

A large number of possib le determinants came alon g in explor ing li terature in

chapt er 2. Obviously , not a l l determinants can be included in our analysis.

Some i t ems can simply not be measured , l ike ‘good luck’. Ot he rs may not be

included in the da tase t . Moreover , some de terminants could be const ruc t ed

but were excluded in the process o f est imat ion . In the fol lowing subsec t ions,

the var iables included are descr ibed in the f ramework of our classi f ica t ion. For

each determinant some br ief explanat ion and the def ini t ion are provided .

M ost de terminants become opera t iona l as dummy var iables.

3.4.1 Human cap ital

Human capita l re lates to the intr insic qual i t ies and is thought to have a posi-

t ive inf luence on the success of star t ing entrepreneurs. Knowledge is an im-

por tant f actor in this respec t , w hich may be acquired through general educ a-t ion or through t ime . Part icular entrepreneur ia l capabi l i t ies are general ly not

taught in the educa t ion programmes. Knowledg e of the wor ld, knowledge o f

business and f inance , and educat ional level may contr ibute to entrepreneur ia l

succ ess. The fol lowin g variables tha t re la te to human cap i ta l could be der ived

from the f irm founders survey and are included in our analysis.



19

Age

The age o f th e ent repreneur can be considered as a measure for knowledge of

the wor ld . In the survey, the responden ts were asked to st a te the age ca tegory

they were in ( intervals of 5 years). The age var iable takes on the value of the

mean of t he age int erval d iv ided by 10. For example , i f an entreprene ur is in

the categ ory 20-25, the age var iable takes on the value of 2 .25 .

E du c at i on

I f an entrepreneur has exper ienced a high level of educat ion (university or

high-level vocat ional t ra ining), the d ummy var iable for educa t ion a t t a ins the

value one (and zero otherwise) . In the survey, people were asked to choose

the highest educat ion level they have comp le t ed.

E xper i en c e i n the s e c tor

This dummy var iable indica t es whether the respondent has exper ience inworking in the same sec tor as he star ted the f irm .

E xper i en c e i n trade

I f the entrepreneur is already famil iar with t rade , th is may enhance his results.

The dum my var iable at ta ins the value of one i f there is some exper ience in

trade , and of zero otherwise .

E xper i en c e a s an emp l oyee

Having exper ience as an employee may be benef icia l to the star t ing entrepre-

neur , as i t wi l l enha nce knowledge o f business. The dummy var iable equals

one i f the respondent has been an em ployee and zero i f th is is not the case.

E xper i en c e i n s e l f-emp l oyment

Firm founders with exper ience in self -employm ent may have be t ter prospec ts

for achieving success. The var iable equals one i f the f irm foun der comes f rom

self -employment, and zero o therw ise .

F i nan c i a l exper i en c e

F inance is an importan t aspec t of entrepreneurship . Some knowled ge on f i -

nancing the business is almost a necessi ty. I f the entrepreneur has some f inan-

cia l exper ience himself , the var iable equals one (and zero otherwise) .



20

Table 6 Descriptive stat istics of determinants con sidering human capital

Variable Mean St d . Dev. Min . M ax.

age 2.40 0.89 2.25 6.75

high education 0.29 0.45 0 1

experience in the sector 0.63 0.48 0 1

experience in trade 0.70 0.46 0 1

experience as an employee 0.94 0.23 0 1

experience i n self-employment 0.08 0.27 0 1

f inancial experie nce 0.08 0.28 0 1

3.4.2 Financial cap ital

Almost every f irm needs a substant ia l f inancia l input . The f inancial resources

o f the star t ing entrepreneur are therefore important . Many quest ions regard-

ing this top ic are posed in the f irm found ers survey . We will use the fo l low ing

determinants.

Amount other i n c ome

I f the f irm founder receives income outside the income from self -employment ,

t his reduces the r isks for the f irm foun der . The amount o f o ther income could

be indicated in six categor ies, ranging from zero to f ive (50 000 Dutch guild-

ers).

Own c ap i ta l

This var iable measures the share - that can be classif ied in 12 categor ies, rang-

ing fro m 0 to 1 (100 percent) - of own capita l in the tota l amount of in i t ia l

capita l that is needed by the ent repreneur .

C ontr i but i on by bu s i ne ss partner ( s )

Maybe a business par tner makes some f inancia l cont r ibut ion. This partner wi l l

(p ossib ly) benef i t f rom the ne w f irm. This wil l again be indica ted using a

dummy var iable .

R e c e i ved l oan from fam il y

I f the family comes to help in the f inancing part of star t ing the f irm , this is in-

dicated by anot her dummy variable .

Wage i n c ome re c e i ved by s pou s e

For determining the succ ess of households that have dou ble income , a var iable

is included tha t de termines whether the spouse receives an income (var iable

equa ls one) or not (variable equals zero).



21

Table 7 Descriptive stat istics of determinants con sidering f inancia l capital

Variable Mean St d . Dev. Min . M ax .

amount oth er income 2.83 1.97 0 5

f inanced with own capital 0.67 0.40 0 1

contribution by busin ess partner(s) 0.11 0.31 0 1

re ceived loan from family 0.17 0.38 0 1

wage income received by spo use 0.44 0.50 0 1

3.4.3 Social cap ital

In order to have succ ess as an ent repreneur, the impor tance o f communica t ing

with re lat ions seems to be growing and growing . Having capabi l i t ies in this re-

spec t can be indica ted as ow n ing social capita l1

. A lso in l i terature , increased

at tent ion is pa id on the re la tion be tw een social capita l and entrepreneurship2

.

The fo l low ing i t ems dea l ing w i th social capita l could be d er ived f rom the f irm

founders survey .

E ntrepreneur s i n the fam il y

Being inf luenced by self -employed members of the family can be a det ermin-

ing fac tor for success. For this, a variable is const ruc t ed tha t has value on e i f

the f irm founder reported considerable inf luence o f self -employed family

members, value 0.5 if there is some inf luence and zero i f th is is not the case .

C onta c t w i th entrepreneur s i n network s

This var iable indica tes whether the f irm founder has ofte n (value eq uals one),

somet imes (value equals 0.5) or never (value e qua ls zero) contac ts with o ther

ent repreneurs.

E mot i ona l s upport from s pou s e

This determinant re la tes to the emot iona l suppor t f rom the spouse . I f this is

presen t , the var iable equals one , and zero o therw ise .

B oard i ng out of a c t i v i t i e s to other s

Does the entrepreneur board out a c t ivit ies? This suggests at least a st ra teg ic

choice made by the entrepreneur . Of t en , th is boarding out is to a re lat ionship

already known to the entrepreneur before the star t -up. I f the entrepreneur

does board out , this variable is equal to one , and i t is zero otherwise .

1 Jacobs (1961) int rod uced this as a term. Sin ce then, there have been several interpretat ions re -garding socia l capital. We w ill use the one proposed by Loury (1977), who defined it as natu-rally occuring relationshi ps to promote or aid the development of valued skills or characteris-tics.

2 See for exam ple Brüderl and Preisendorfer (1998), Cooke and Wills (1999).



22

Table 8 Descriptive stat istics of determinants considering socia l capital

Variable Mean St d . Dev. Min . M ax .

inf luenced by entrepr. in the family 0.26 0.35 0 1

contact with entrepr. in networks 0.28 0.36 0 1

emotional support from spo use 0.73 0.44 0 1

boarding out of act ivit ies to others 0.45 0.50 0 1

3.4.4 Keeping up w ith developments relevant for the businesses

How do the en t repreneurs get the i r in format ion on th e deve lopments tha t a re

importa nt for their part icular businesses? A number of que st ions regard ing

this top ic have been posed to the ent repreneurs. These were the fol lowing:

• Do you

− visi t congresses

− take part in courses

− keep up w i th l i t e ra ture− get informat ion f rom customers

− get informat ion f rom suppl iers

− get informat ion f rom banks

− get in format ion f rom commerci al coopera t ion

− get informat ion f rom branch organisa t ion

− get informat ion f rom fe l low-entrepreneurs?

Each respondent answered 1 (of ten), 2 (somet imes) or 3 (never). A fac tor

analysis on the answ ers to the quest ions revealed the fol lowing1

. Three fac tors

were found w i th load ings greater tha n 1 (cumulat ive account ing f or 52.5% of

the var iance) . A four th f ac tor (loading 0.9) is added such that 61.5% of thevar iance is accounted for. These four fac tors can be given the fol lowing inter-

pre ta t ions as given in Table 9. The fac tors can be seen as potent ia l success de-

terminants and w il l be included in our analysis in chapter 4.

The subjec t of keeping up with business is closely re lated t o soci al cap i ta l.

However , i t cannot be considered as a component o f socia l capita l . I t ref lec ts

t he st rategy used to retr ieve re levant informat ion f rom re lat ion ships. These re-

la t ionships do not occur natura l ly.

Table 9 Factor analysis regarding quest ions referring to ‘keeping up with busin ess’

Direct ion of focus to keep up Variables with loadings greater than 0.5

1. Focus on branch in general Branch , congress, courses, l i terature

2. Fo cus on direct bu sin ess re lat ions Cu stomer, supplier

3. Focus on commercial relat ions Banks, commercia l cooperat ion

4. Fo cus on fellow-entrepreneurs Contact with fel low- entrepreneurs

3.4.5 Controls

1 Principal components with varimax rotat ion, missing values ar e set to m ean (excluding thesedoes not lead to signif icant changes).



23

In this sec t ion we deal with potent ia l determinant s t ha t could be important in

determining entrepreneur ia l succ ess, that do no t f i t in the cl assi f ica t ion . In clu-

sion is necessary , as neglec t ing these var iables could result in biased est ima-

t ions.

Gender

Gend er is known for a l l entrepreneurs in the survey. The dum my variable

equa ls one i f the respondent is male , and zero i f the respondent is f emale .

Goa l : emp l oyment growth

Some ent repreneurs simply do not wan t to hire employee s. These entrepre-

neurs are appointe d a value of zero, while the others have values one to

three, depending on the repor ted de termina t ion to h i re employees.

F u ll s e l f-emp l oyment

I f the entreprene ur is not ful ly employed , th is can have i ts e f fec ts on per form-

ance of the f irm . A d ummy var iable is crea ted tha t a t t a ins a value of one i f the

respondent ind ica t es tha t he is self-employed for 100 percen t .

S e c tor dummy bu s i ne ss s erv i c e s

Business services may need speci a l a t t ent ion . Entrepreneurs who are ac t ive in

the business services are given a value of one , while a l l other entreprene urs

have a value of zero .

P u ll mot i ve to be c ome s e l f-emp l oyed : h i gher i n c ome than wage s

A var iable is construc t ed to study the inf luence of the mot ive based on higherexpec t ed ga ins on performance . The var iable at ta ins a value of one i f th is is

st rongly the case , o f 0.5 if this is moderate ly the case and of z ero i f th is is no t

the case .

Table 10 Descriptive stat istics of determinants considering the control variables

Variable Mean St d . Dev . Min . M ax .

gender 0.72 0.45 0 1

goal: employment growth 0.40 0.35 0 1

100 percent self-employed 0.93 0.26 0 1

sector dummy: busin ess services 0.19 0.39 0 1

motive to becom e self-empl.: higher income 0.39 0.39 0 1

3.5 Discussion

This chapt er provided a descr ipt ion of the dataset and a classi f icat ion of the

construc ted determinant var iables. Unfor tunate ly, some possib le de terminants

(such as ‘good luck’) simply cannot be mode l led . At t he same t ime , choices

have to be made: the data se t conta ins many more opportunit ies for construc t -



24

ing o ther , dif ferent determinants. The approach of the und er lying study has a

genera l charac t er. Therefore , using the classi f icat ion appl ied , the most appea l -

ing de terminants were considered. When focussing on speci f ic top ics, it will be

possib le to study the inf luences of o ther, more speci f ic determinants.



25



26

4 Empirical results

4.1 Introduction

In order to test the impac t of a number of determinant s on entrepreneur ia l

success some mode ls are construc t ed tha t incorporate a l l general ef fec ts of the

success determinants1

. The concept of non-response may be importa nt in de-

termining the re lat ionships between de terminants and success. This matter is

dealt with in sec t ion 4.2. The models, as well as their results, are descr ibed in

the next three sect ions for the succ ess measures prof i t , employment and dura-

t ion . Whereas the mode ls for prof i t and employmen t are est imate d using or-

dinary least squares, a survival model is construc ted for durat ion .

4.

2 Non-responseThe percentage of f irms report ing failure in 1995 is only 6 percen t , and 20 per-

cent in 1997. Earlier re search - conduc t ed in the Nether lands - show ed fa i lure

rates of 10 percent in the f irst year and of 26 percent in the f irst three years,

see EIM (1994). This suggests tha t the percenta ge of fa i l ing f irms in the group

that did not respond to the la te r quest ionnaires is substant ia l ly higher than

the percenta ge of fa i l ing f irms that did provide performance in format ion . The

proba bi l i ty dist r ibut ion of performance for the responden ts with comple te

da ta is dif ferent f rom th e probabi l i ty dist r ibut ion of performance in the popu-

la t ion. This causes a comple te da ta sample selec t ion bias.

F inancia l per formance may inf luence the abi l i ty or wi ll ingness of a responden tto f i l l out a quest ionnaire . Whe n a f irm has stopp ed a l l act ivit ies i t is of ten not

possib le to con t ac t the f irm to conduc t a poll. Keasey and Watson (1991) men-

t ion this problem as the major hindrance for gath er ing small f i rm performance

da t a . Even if a faile d f irm is reached by the conduc tors of the survey , i t may be

tha t the respondent does not see the use of f i ll ing out the qu est ionnaire since

the f irm does not exist any more . Another possibil ity is that th e responden t

simply does not fee l l ike report ing fa i lure .

Besides the mode ls that are dealt with in the next sec t ions, models were con-

struc ted th at al low for a re lat ionship be tween response to the quest ionnaire

in 1997 and p erforman ce.

I t was

hypothes

is

ed that the wi l l ingness

to res

pondw as posi t ive ly re lated to p erformance . These mode ls did not outperform the

mode ls without non-response .

1 The ef fects of the determinants for the dif ferent function aspe cts (as mentioned in chapt er 2)did not dif fer signif icantly. Therefore, we only discuss the resul ts of the general model thatdoe s not dist inguish the e ntrepreneurial funct ions.



27

4.3 Profit

Le t πi be the pro f i t for responden t i in 1997, and x ij respondent i ‘s value of de-

terminant j in 1994. The model to be es t imated is the fo l low ing:

iij

J

j

j i x εβαπ ++= ∑=1

ln , where ),0(~ 2 pi N σε .

We have speci f ied the logar i thm of prof i t as the de pendent var iab le ra ther

than prof i t i tself , because we think changes in the de terminants in f luence

relat ive profit ( in percen t ages) rather tha n absolute pro f i t (in gui lders). The

equat ion above can be est imated using ordinary least squares. Results are

show n in Table 11.

Human capita l var iables appear impor tant de terminants of the prof i t of an

ent repreneur. Four var iables appear to be signif icant . Having exper ience in

t he sec tor appears to have the largest in f luence: entrepreneurs with experi-

ence in the sec tor earn about thr ice as much compared to en t repreneurs lack-ing th is experience . A lso , the in f luence of high educat ion and exper ience in

self -employme nt is substant ia l : having a high educat ion approximately dou-

bles prof i t , the same holding t rue for having exper ience in self -employment ,

a l though the uncerta inty about th e right va lue of this l a t t e r in f luence is lar-

ger. Age appears to have a negat ive inf luence on prof i t : an entrepreneur th at

is 10 years older at the start of his business earns abou t 25% less than his

younger equivalent .

As regards f inancia l capita l var iables, i f the entrepreneur receives anot her in-

come beside the prof i ts der ived f rom self -employment , th is reduces prof i tabi l-

i ty. Every 10 tho usand gui lders he ge ts f rom other sources comes toge therwith about 15% lower prof i ts. The other f inancia l capita l var iables appear to

have no signif icant inf luence .

As regards socia l capita l determinants, the emot iona l suppor t o f the spouse

appears important : entrepreneurs having this suppor t make abou t 80% more

prof i t than their counterpar ts having to do without this suppor t . How eve r,

contrary to wh at one would expe c t , get t ing he lp and feedback f rom se l f -

employed members of the fa mily inf luen ces prof i ts st rongly negat ive . Those

who ge t th is help make ab out a third of the prof i ts made by their counter-

parts doing without this he lp and feedback .

St rategies for keeping up with business that are l inked to prof i ts are thest rategies tha t focus on direc t business and commercia l re lat ions.

Finally , neither of the control var iables appears to have a signif icant in f luence

on prof i ts.



28

Ta ble 11 Est imation resul ts for profit (logarithm)

est imatea standard error

human capital

age -0.264 0.135 **high education 0.721 0.258 **

experience in the sector 1.059 0.254 **

experience in trade 0.196 0.268

experience as an employee -0.204 0.407

experience i n self-employment 0.752 0.426 *

f inancial experience 0.144 0.392

financial cap ital

amount oth er income -0.183 0.069 **

f inanced with own capital -0.265 0.297

contribution by bus

in ess

partner(s

) 0.

560 0.

365re ceived loan from family 0.174 0.303

wage income received by spo use 0.041 0.263

social capital

inf luenced by entrepr. in the family -1.024 0.318 **

contact with entrepr. in networks -0 .053 0.317

emotional support from spo use 0.582 0.259 **

boarding out of activities to others 0.150 0.228

keep up strategy variables

focus on branch in gen eral -0.162 0.116

focus on direct bu sin ess relat ions 0.234 0.112 **focus on commercial relat ions 0.213 0.116 *

focus on fel low-e ntrepreneurs 0.187 0.112

control var iab les

gender 0.323 0.259

goal: employment growth 0.443 0.331

100 percent self-employed 0.467 0.380

sector dummy: busin ess services 0.171 0.292

motive to becom e self-empl.: higher income 0.397 0.293

number of observat ions 998

log likelihoo d -1728.60

a As the logarithm of prof it is modelled , an e st imate wit h valueβ , means that if the determining(dummy) variable changes from 0 to 1, the prof it changes with a factor exp(β).

** Significant at 5% level.

* Significant at 10% level.



29

4.4 Emp loyment

The model for employment is ana logous to the mod el for prof i t . Let Li be the

cumula ted employment gen era ted by responden t i bet wee n 1994 and 1997,

and x ij respondent i ‘s va lue o f de terminant j in 1994. The model to be est i-

mated is the fo l low ing:

e

iij

J

j

j i x L εβα ++= ∑=1

ln , where ),0(~ 2

e

e

i N σε .

Re sults are depic ted in Table 12.

Contrary to what was the case when ana lysing the de terminants of prof i ts,

human capita l var iables app ear to be not dominan t in expla ining the employ-

ment genera ted by ent repreneurs. Only two var iables have signif icant inf lu-

ence . Having f inancia l exper ience seems to have a posi t ive inf luence: entre-

preneurs having this exper ience generate near ly twice the employment com-

pared to their counterpar ts lack ing th is experience . A ge , on the contrary, l eads to less employmen t : i f an entrepren eur is 10 years older at the star t he gener-

a tes about 40% less employment.

As regards f inancial capita l var iables, i t appears that the more is done w i th

ou tside capita l the higher the generated employment is. I f business has been

star ted with 10% point more out side capita l th is l eads on average to 7% more

employment . Furthermore , ent repreneurs benef i t ing f rom a f inancia l contr i-

but ion by a business partner generate about two and a half a s much employ-

ment as their counterpar ts doing w i thout such a cont r ibut ion. I f the entrepre-

neur has other income avai lable besides what he earns f rom his business, t his

has a negat ive impac t on the employment genera ted .

As regards socia l capita l var iables, ent repreneurs having con t ac ts with o ther

ent repreneurs in ne tworks, genera te about tw ice as much employment as

their counterpar ts lack ing these ne tworks. Furthermore , ent repreneurs board-

ing out ac t ivit ies genera te about 50% more employment .

St rategies for keeping up with business that are l inked to gen erat ing em-

ployment are the st rategie s tha t focus on commercia l re lat ions and fe l low-

ent repreneurs.

Tw o control var iables appear to be signif ican t . As w as to be expec t ed , ent re-

preneurs who had as goal r ight f rom the star t to generate employment

growth appear to have succeeded in this respec t when compared t o their fe l-

low-entrepreneur s who did not have this in tent ion. On average , they appear

to have created e ight t imes as much employment . Curiously , t hose who are

full-t ime self -employed, genera te less employment than their part - t ime coun-

terparts: about 60% less.



30

Tab le 12 Estimation resul ts for cumulated employment (logarithm)


human capital

age -0.533 0.107 **

high education 0.105 0.211

experience in the sector 0.045 0.208

experience in trade -0.298 0.219

experience as an employee -0.127 0.338

experience i n self-employment 0.525 0.336

financial experience 0.614 0.331 *

financial cap ital

amount oth er income -0.162 0.055 **

f inanced with own capital -0.729 0.236 **

contribution by busin ess partner(s) 0.884 0.286 **

re ceived loan from family 0.134 0.242

wage income received by spo use 0.344 0.212

social capital

inf luenced by entrepr. in the family -0.174 0.252

contact with entrepr. in networks 0.815 0.254 **

emotional support from spo use 0.022 0.206

boarding out of activities to others 0.425 0.183 **


focus on branch in general 0.031 0.093

focus on direct bu sin ess relat ions 0.112 0.089

focus on commercial relat ions 0.197 0.091 **

focus on fel low-e ntrepreneurs -0 .154 0.088 *

control var iab les

gender -0.121 0.208

goal: employment growth 2.064 0.266 **

100 percent self-employed -0.975 0.299 **

sector dummy: busin ess services -0 .199 0.236

motive to becom e self-empl.: higher income 0.070 0.235


log likeliho od -1209.31

a As the logarithm of generated employment is modelled , an e st imate with valueβ , means thatif the de terminin g (dummy) variable changes from 0 to 1, the generated employment

changes with a factor exp(β).





31

4.5 Duration

For durat ion , we apply a dif ferent model than for prof i t and employment , vi z .

a survival model . In such a mod el the (logar i thm of t he) expec t ed survival t ime

is model led as a func t ion of the charac terist ics of the ent repreneurs. A descrip-

t ion of this model is provided in the appendix . Re sults are show n in Table 13.

Po ssessing human capita l appears to be important for the durat ion of the

business. Three human capita l- re lated determinants appear to have a signif i-

cant posi t ive inf luence on durat ion . Star t ing at an age ten years higher in-

creases the expec ted durat ion by about one third . Having exper ience in the

sector or as an employee increases the expec ted durat io n by about 70 perce n t .

The in f luence of f inancia l exper ience has the unexpect ed sign: it seems to af -

f ec t survival negatively . Having f inancia l exper ience decreases the expec t ed

durat ion by approximately 40 percen t .

F inancia l capita l does no t seem to play any role in expla ining the durat ion of a

f irm .

Social capita l af fec ts survival through the emo t ional suppor t o f the spouse ( the

pre sence of this aspec t raises expec t ed survival t ime by around 50 percent),

and through the boarding out of ac t ivit ies to o thers (which is expec ted to in-

crease survival t ime by about on e third as compared w i th ent repreneurs w ho

do not board out).

St rategies of keep ing up w i th business seem to be of importance , with e f f ec ts

coming f rom the focus on the branch in general and commercia l re lat ions.

In contrast to the measures of success dealt with before , gender does make a

dif ference: male entrepreneurs have signif icant ly lower hazard rates and arethus considered to be more successful as far as durat ion o f the f irm is con-

cerned . The associ a t ed advantag e expressed in re lat ive expec t ed survival t ime

is est imate d rather high at about 50 percen t . Whether the ent repreneur is self-

employed ful l-t ime is the only other control var iable that is o f signif icance for

expla ining the durat ion of the f irm . An ent repreneur who is self -employe d for

100 percent has lower probabi l i t ies to quit (low er hazard rates) and is ex-

pec ted to have a survival t ime that is approximately 1.5 t imes higher than tha t

of part - t ime entrepreneurs.



32

Tab le 13 Est imation resul ts for duration (logarithm)


human capital

age 0.273 0.097 **high education 0.085 0.190

experience in the sector 0.547 0.177 **

experience in trade 0.295 0.183

experience as an employee 0.525 0.265 **

experience i n self-employment -0.454 0.288

financial experience -0 .525 0.289 *

financial cap ital

amount oth er income -0.034 0.050

f inanced with own capital -0.076 0.207

contribution by bus

in ess

partner(s

) 0.

146 0.

258re ceived loan from family -0.003 0.215

wage income received by spo use -0 .239 0.189

social capital

inf luenced by entrepr. in the family -0.078 0.220

contact with entrepr. in networks -0 .169 0.228

emotional support from spo use 0.407 0.177 **

boarding out of activities to others 0.295 0.163 *


focus on branch in general 0.237 0.081 **

focus on direct bu sin ess relat ions -0 .027 0.079focus on commercial relat ions 0.228 0.091 **

focus on fel low-e ntrepreneurs 0.085 0.079

control var iab les

gender 0.409 0.177 **

goal: employment growth 0.080 0.231

100 percent self-employed 0.928 0.250 **

sector dummy: busin ess services -0 .198 0.208

motive to becom e self-empl.: higher income -0.228 0.208


log likeliho od -1141.30

a As the logarithm of expect ed survival time is modelled, an e st imate with valueβ , means that if thedet ermining (dummy) variable changes from 0 to 1, the expect ed survival time changes with a factor

exp(β).





33



34

Literature

Brüder l, J. and P. Preisendörfer (1998), Ne t w ork Suppor t and the Success o fNewly Founded Businesses. S mall Bu s ine ss E c onomi c s 10: 213-225.

Cant i l lon , R. (1931), E ss ai s ur la Nature du Commer c e en Général , ed i t ed andtransla ted by H . Higgs, London: Macmil lan.

Cooke, P. and D . Willis (1999), Small Firms, Socia l Capita l and the Enhancemen tof Business Performan ce Through Innovat ion Programmes, S mall Bu s ine ss E c o-

nomi c s 13:219-234.

Di jk , B. van and A .R. Thurik (1995, in Dutch) , Ent repreneurship: visies en bena-der ingen , EIM Re s e ar c h Report 9510 / N , Zoetermeer: EIM .

Di jk , B. van (1996), Determinant s of S u cc e ss fu l Entrepreneur s hip , d oc tora lstudy, Rotterdam: Erasmus Universi ty .

Greene , W .H . (1997), E c onomet r i c Ana ly s i s , Prentice-Hal l, In c .

Héber t , R.F. and A .N . Link (1988), The Ent repreneur , M ain s t ream View s and

Radi c al Crit ique s , Praeger, New York .

Jacobs, J. (1961), The Li f e and De ath o f Grea t A mer i c an Cit ie s , New York : Ran-dom House .

Keasey, K . and R. W atson (1991), The State of the Art of Small Firm FailurePredic t ion: Achievements and Prognosis, In te rna t iona l S mall Bu s ine ss J ourna l , Vol. 9, pp . 11-28.

Kirzner , I. M . (1979), Pe rcept ion , Oppor tun i t y, and Prof i t : Stud ies in the Theoryof Entrepreneurship , Chicago: U niversity of Chicag o Press.

Kirzner , I. M . (1981) Uncerta inty , Discovery and Human Ac t ion, paper pre-

sented at the New York Universi ty-Liberty Fund Centenary Conference onLudwig von Mises, New York .

Kn ight , F. H . (1921), Ri s k , Un c erta inty a nd Prof i t , New Y ork : Houghto n Mi f f l in.

Loury, G . (1977), A Dynamic Theory of Racial Income Di f f e rences, in P. Wal laceand A . La Mond (eds.), Wom en , Minor it ie s and Employment D i s c r iminat ion , Lexington , MA: Hea l th , pp . 153-186.

M arshal l, A . (1961), Prin c ip le s o f E c onomi c s , London: Macmil lan .

Menger, C. (1950), Prin c ip le s o f E c onomi c s , t ransl a t ed by J. Dingwall and B. F . Hosel i tz, Glencoe , Il l .: Free Press.

Ne ther lands Ministry of Economic Af fa i rs (1999, in Dutch) , De ondernemende

s amenleving . Meer kan s en , minder bedre ig ingen voor ondernemer s c h ap , Gov-ernment W hite Paper, The Hague .

Praag , M . van , 1996 , Determinant s o f S u cc e ss fu l Ent repreneur s hip , A mst e rdam:

Thesis Publishers.

Praag , C. M . van (1999), Some Classic View s on Ent repreneurship , D e E c onomi s t

147 : 311-335.

Say, J. B. (1845), A Treat i s e on Polit i c al E c onomy , Philadelphia: Grigg & Elliot .

Schumpeter , J. A . (1934), The Theory of E c onomi c Deve lopment , Cambr idge , M ass. : Harvard U niversity Press.

Schumpeter, J. A . (1942), Capita l i s m , S o c ial i s m and Demo c ra c y , New York:Harper and Row .



35



36

Appendix The survival model

In thes

urvival model,

f i rm founders

s

tar t their bus

iness

at t ime t =0.

Exits

aredescr ibed by a random process governed by a probabi l i ty densi ty func t ion f (t )

and adjoining dist r ibut ion func t ion F(T). Thus, the pro babi l i ty of a f irm exit -

ing be t w een t and t +dt is denoted by f(t)dt, and th e l ike l ihood that a f irm exits

in the f irst T months is denoted by F(T).

Consequent ly, the survivor func t ion S(T), def ined as the proba bi l i ty that a

f irm survives the f irst T months, is equivalent to 1-F(T). A lso , the hazard rate

θ(t ), speci fy ing the condit ion al probabi l i ty that a f irm that has remained in

business for a period from 0 to t , exits in the short interval [t, t+dt ], is equiva-

lent to:

(1) )(

)(

)( t S

t f

t =θ .

The hazard func t ion is model led as a func t ion of a set of exogenous person -

speci f ic regressors, the vec t or x , and of t ime t to permit durat ion dependence .

Assuming the absence of regressors, the hazard is a non-monoton ic f unc t ion

of t . The assumpt ion was show n to hold by a f irst inspec t ion of the durat ion

data . A simple hazard speci f icat ion that permits non-monoton ic behaviour is

the log-logist ic (see Lancast er, 1992, p . 44),

(2)α

ααθ

t xk

t xk xt

)(1

)(),(

1

+

=

−

.

Le t )'exp()(k β= . Then, i t can be demonst ra ted tha t

(3) ii x xt E 'ln β−=1

.

These are the coef f ici en ts (i . e . -β) that are show n in Table 13. The l ike l ihood

func t ion to be maximize d is the fo l low ing:

(4) id

iiii t t S L )()( θ= ,

where id =1 i f indiv idual i’s exit i s observed at i

t and id =0 if i’s lengt h of time

in business is r ight censored . As equa t ion (1) can be rewrit ten as

)()(ln)(

t d t S d t −=θ , i t fo l low s t ha t

−= ∫

t

duut S 0

)(exp)( θ . The logar i thm of

the l ike l ihood func t ion can thus be wr it ten in terms of the hazard func t ion as:

(5) )(ln)(ln0

iii

t

ii t d duu L

i

θθ +−= ∫ .

1 See Greene (1997), page 993.



37



38

List of Research Reports

The research report series is t he successor of both the research pap er and th e 're-

searchpubl i

cat ie ' series. There is a

conse

cut ive repor t number ing fo l low ed by / x . For

/ x there are f ive opt ions:

/ E: a report of the business uni t St ra teg ic Re search , writ ten in Engl ish;

/ N: like / E, but wr it ten in Dutch;

/ F: like / E, but w r it ten in French;

/ A: a report of one of the other business units of EIM / Small Business Re search

and Consul tancy;

/ I : a repo rt of the business unit St rategic Re search for internal purposes; exter-

nal avai labi l i ty on request .

9301 / E The inter tem poral stabi l i ty of the concentrat ion-margins re la t ionship in

Dutch and U .S. manufac tur ing; Yvonn e Pr ince an d Roy Thur ik

9302 / E Persist ence of prof i ts and compet i t iveness in Dutch manufac tur ing; Aad

Kle i jweg

9303 / E Small store presence in Japan; Mart in A . Carree, Jeroen C. A . Pot jes and A .

Roy Thurik

9304 / I Mult i-fac torial r isk analysis and the sensit ivity concep t; Erik M . Vermeulen ,

Jaap Spronk and Nico van der Wijst

9305 / E Do small f irms' price-cost margins fo l low those of large f irms? First empiri-

cal resul ts; Yvonn e Prince an d Roy Thurik

9306 / A Export success of SMEs: an empirical study; Cinz ia Ma ncin i and Yvon ne Pr in-

ce9307 / N Het aandeel van het midden- en k le inbedr i j f in de Neder landse industr ie;

Kees Bakker en Roy Thur ik

9308 / E Mult i-fac torial r isk analysis appl ied t o f irm evaluat ion; Erik M . Vermeulen ,

Jaap Spronk and Nico van der Wijst

9309 / E Visual iz ing inter f irm compar ison; Erik M . Vermeulen, Jaap Spronk and Nico

van der Wijst

9310 / E Industry dynamics and small f i rm developme nt in the European pr int ing

industry (Case Stud ies of Br i ta in , The Nether lands and D enmark) ; Michae l

Kitson , Yvonne Prince and Met te Mönst ed

9401 / E Employmen t during the business cycle: evidence from Dutch manufacturing;

Marcel H.C. Lever and Wilbert H. M . van der Hoeven

9402 / N De Neder landse indust r ie in interna t ionaal perspec t ie f : arbeidsproduk-

t iv i te i t, lonen en concurrent iepo si t ie ; Aad Kle i jweg en Sj aak Vo l lebregt

9403 / E A micro-econometric analysis of interrelated factor deman d; René Huigen , Aad

Kleijweg , George van Leeuw en and Kees Zeelenberg

9404 / E Between economies o f scale and entrepreneurship; Roy Thurik

9405 / F L'évolut ion struc ture l le du commerce de gros f rançais; Luuk Klom p et Eu-

gène Rebers



39

9406 / I Basisinkomen: een inventar isat ie van argume nten; Bob van Di jk

9407 / E Inter f irm performa nce evaluat ion under uncerta inty , a mult i-dimensiona l

f rame-work; Jaap Spronk and Erik M . Vermeulen

9408 / N Indicatoren voor de dynamiek van de Neder landse economie: een sec tora le

ana lyse; Garmt Dijkst e rhu is, Hendr ik-Jan Heeres en Aad Kle i jweg9409 / E Entry and exit in Dut ch manufac tur ing industr ies; Aad K le ijweg and Marcel

Lever

9410 / I Labour produc t iv i ty in Europe: dif ferences in f irm-siz e , countr ies and indus-

tr ies; Garmt Dijksterhuis

9411 / N Verslag van de derde mondia le workshop Small Business Economics; Tin-

bergen Inst i tuut , Rotterdam , 26-27 augustus 1994; M . A . Carree en M . H.C.

Lever

9412 / E Internal and external forces in sec tora l wage forma t ion: ev idence f rom the

Nether lands; Johan J. Graaf land and Marcel H .C. Lever

9413 / A Sel ectie van leverancie rs: een kwest ie van produkt , profijt en partnerschap?; F.

Pleijster

9414 / I Grafische weergave van tabellen; Garmt Dijksterhuis

9501 / N Over de toe passing van de f inancieringsthe orie in het midde n- en klei nbe drijf;

Erik M . Vermeulen

9502 / E Insider power, market power, firm size and wages: evidence from Dutch manu-

facturing in dustr ies; Marcel H .C. Lever and Jolanda M . van Werkhooven

9503 / E Export performa nce of SMEs; Yvonne M . Prince

9504 / E Strategic Niches and Profitability: A First Report; David B. Audretsch, Yvonne M .

Prince and A . Roy Thurik

9505 / A Me er over wink elope nstellingstijden; H.J. Gianotten en H.J. Heeres

9506 / I Interstratos; een onderzoek n aar de mogelijkheden van de Interstratos-dataset;

Jan de Kok

9507 / E Union coverage an d sectoral wages: evidence from the Netherlands; Marcel

H.C. Lever and Wessel A . Marquering

9508 / N O nt wikkeling van de grootteklassenverdeling in de Nederlandse Industrie;

Sjaak Vollebregt

9509 / E Firm size and employment det ermination in Dutch manufacturing industries;

Marcel H .C. Lever

9510 / N Entrepreneurship: visies en ben aderinge n; Bob van Dijk en Roy Thuri k

9511 / A De toegevoegde waarde van de detailhandel; enkele verklarende theorieën

tegen de achtergrond van ontwikkelingen in distr ibutiekolom, technologie en

externe omgeving; J.T. Nienhuis en H.J. Gianotten9512 / N Haalbaarheidsonderzoek MA NA GEMENT-model; onderzoek naar de mogelijk-

heden voor een simulatiemodel van het bedrijfsleven, gebaseerd op gedetail-

leerde branche- en bedrijfsgegevens; Aad Kleijweg, Sander Wennekers, To n

Kwaak en Nico van der Wijst

9513 / A Chipp en in binn en- en buitenlan d; De elektronische portemonnee in kaart ge-

br acht; een verkenning van toe passingen, mogelijkheden en consequenties va n



40

de chi pcard als elektronische portemonnee in binnen- en buitenland; drs. J. Ro-

orda en drs. W .J.P. Vogelesang

9601 / N Omzetprognoses voor d e d etailh and el; Pieter Fris, Aad Kleijweg en Jan de Kok

9602 / N Flexibiliteit in de Ned erlandse Industrie; N.J. Reincke

9603 / E The Decision be tw een Internal and External R&D; David B. Audretsch, Albert J. Menkveld and A . Roy Thurik

9604 / E Job creat ion by si ze class: measurement and empir ical invest iga t ion ; Aad

Kle ijweg a nd Henry Nieuwenhui jsen

9605 / N Het e f f ec t van een beursnoter ing; drs. K .R. Jonkheer

9606 / N Een Micro-werkge legenhe idsmodel voor de Deta i lha ndel; drs. P. Fr is

9607 / E Dema nd for and wag es of high- and low-ski l led labour in the Nethe r lands;

M .H .C. Lever and A .S.R. van der Linden

9701 / N Arbe idsomst and igheden e n bedr i jfsgroot te . Een verkennin g met de LISREL-

methode; drs. L.H . M . Bosch en drs. J. M .P. de Kok

9702 / E The impa c t o f compe t i t ion on pr ices and wages in Dutch manufac tur ing

industr ies; Marcel H .C. Lever

9703 / A F A M OS, een f inancier ingsmodel naar groot tek lassen; drs. W . H .J. Verhoeven

9704 / N Banencreatie door MKB en GB; Pieter Fris, Henry Nieuwenhuijsen en Sjaak Vol-

lebregt

9705 / N Naar een bedr i j fst ypenmode l van he t Ned er landse bedr i j fsleven; drs.

W . H . M . van der Hoeven , drs. J. M .P. de Kok en drs. A . K w a a k

9801 / E The Knowledg e Soci e t y, Ent repreneurship and Unemployment ; David B.

Audre tsch and A . Roy Thurik

9802 / A Firm Fai lure and Indust r ia l Dynamics in the Nether lands; David B. Aud-

re tsch , Patrick Houwel ing and A . Roy Thurik

9803 / E The determina nts of employmen t in Europe , the U

SA and

Japan; André van

St el

9804 / E PRISM A'98: Policy Re search In st rument for Size-aspec ts in Macro-economic

Ana lysis; Ton Kwa ak

9805 / N Banencreat ie bi j het Kle in-, Midden- en Grootbedr i j f ; Henry Nieuwenhui-

jsen , Ben van der Ei jken en Ron van Di jk

9806 / A Mil ieumo del; drs. K .L. Bangma

9807 / A Barriers for hir ing personnel; Jacques Niehof

9808 / A Methodiek kost en en ba ten Arbowe tgev ing; drs. K . M .P. Brouwers, dr. B.I.

van der Burg , drs. A .F. M . Nij sen en ir. H .C. Visee

9809 / E Business O w n e rship and Economic Grow th; An Empirical Invest igat ion; Mar-

t in Carree , André van St e l, Roy Thurik an d Sander Wennekers 9810 / E The Degree of Col lusion in Construc t ion; M. H .C . Lever, H .R. Nieuwenhui jsen

and A .J. van St el

9811 / E Self -employme nt in 23 OECD count r ies; Ralph E. Wildeman , Geer t Hofst ede ,

Nie ls G . Noorderhaven, A . Roy Thurik , Wim H .J. Verhoeven and A lexander

R. M . Wenneke rs

9812 / E SICLASS: Forecast ing the Europe an enterpr ise sec tor by industry and size

cl ass; Niels Bosma and Ton Kwaak



9901 / E Scanning the Future of Entrepreneurship; drs. N .S. Bosma , drs. A .R. M . Wen-

nekers and drs. W .S. Zwinke ls

9902 / E Are Small Firms Really Sub-optimal?; Compensat ing Fac tor Dif ferent ia ls in

Small Dutch Manufac turing Firms; David B. Audre tsch , George van Leeu-

w en , Bert Me nkveld and Roy Thur ik9903 / E F A M OS; A size-cl ass based f inancia l analysis model; W .H .J. Verhoeven and

E. A . van Noort

9904 / E Conduc t and Performance in Dutch Manufac tur ing; An Appl icat ion of Ap-

pelbau m 1982 with a Plausibil ity-Check; Frank A . Hindr iks, Hen ry R. Nieu-

wenhui jsen and Adr iaan J. van St el

9905 / E Non-compet i t ive Rents in Dutch Manu f ac tur ing; Cond uc t and Per formance

in the Ne w Empirical Indust r ia l Organizat ion; Frank A . Hindriks

9906 / E A human -resource-based theory of the small f irm; Charlotte Koch and Jan

de Kok

9907 / N Van werknem er naar ondern emer; Een hybr ide of direc t e star t?; ir. H .C. Vi-

see en drs. W .S. Zwinke ls

9908 / E Mod el l ing returns to R&D: an a pplica t ion on si ze e f f ec ts; Peter Brouw er

and Henry Nieuwenhui jsen

9909 / E Turbulence and produc t iv i ty in the Net her lands; Niels Bosma and Henry

Nieuwenhui jsen

9910 / E Start-up capita l : Dif ferences betw een male and female ent repreneurs.

‘Does gen der mat ter?’ ; Ingr id Verhe ul and Roy Thur ik

9911 / E Mod el l ing Business O w n e rship in the Nether lands; Niels Bosma , Sander

Wenneke rs, Gerr i t de Wit an d Wim Zwinkels

9912 / A Measur ing innovat ive intensity: Scale construc t ion; J.P.J. de Jong

9913 /

A Determinants of f irm size; Y . Bernardt and R. Muller0001 / E St rategie s, uncerta inty and performance o f small business st a r tups; Marco

van Ge lderen , Michael Frese an d Roy Thur ik

Non Refereed Book Publication 525 Full Text

Documents