FREE ELECTRONIC LIBRARY - Abstracts, online materials

Pages:   || 2 | 3 | 4 |

«Pecuniary and Non-Pecuniary Aspects of Self-Employment Survival Yannis Georgellis*, John Sessions**, and Nikolaos Tsitsianis*** * Department of ...»

-- [ Page 1 ] --

Pecuniary and Non-Pecuniary Aspects of Self-Employment Survival

Yannis Georgellis*, John Sessions**, and Nikolaos Tsitsianis***

* Department of Economics and Finance

Brunel University


Middlesex UB8 3PU


**Department of Economics

University of Bath

Bath, BA2 7AY



Department of Economics

University of Hertfordshire


Hertfordshire AL10 9AB


Abstract: We examine the factors that determine self-employment duration in Britain, paying particular attention to self-reported job satisfaction variables and non-pecuniary aspects of self-employment. Based on spell data from the British Household Panel Study, we estimate single-risk and competing-risks hazard models, separately for males and females. Our results show that job satisfaction is indeed a strong predictor of self-employment exit, even after controlling for standard economic and demographic variables. When five domain job satisfaction measures are used, we find that pay, job security and initiative are the three aspects of self-employment most valued by the self-employed themselves. Gender differences regarding the determinants of self-employment survival and exit destination states are also evident.

Keywords: Self-employment, Survival, Job Satisfaction JEL Classification: J23, J28, J60 Word Count: 5844 Electronic copy of this paper is available at: http://ssrn.com/abstract=941154 I. Introduction Compared to the plethora of studies examining self-employment entry, relatively few studies examine the determinants of self-employment survival and exit. Noteworthy recent examples of empirical studies focusing exclusively on self-employment survival include Johansson (2001) and Taylor (1999). Johansson (2001) uses Finnish longitudinal data and finds that the young, the more educated, and those with previous unemployment experience face a higher risk of exiting selfemployment. Based on data from the British Household Panel Survey (BHPS), Taylor (1999) finds that approximately 40 per cent of self-employment ventures in Britain that started in 1991 did not survive their first year in business. Interestingly, a substantial proportion of self-employment spells is not terminated through bankruptcy but rather through moves to alternative employment. Taylor also highlights the importance of previous unemployment experience, whether individuals quit the previous job, and initial capital as important determinants of self-employment survival. Bates (1990) provides some of first evidence on the probability of surviving in business for a sample of male entrepreneurs in the US and finds that those who possess greater business acumen, labour market skills, and capital, are more likely to survive ceteris paribus.1 Failure to address entrepreneurial success and self-employment survival could cast doubts on the efficiency of government programmes designed to facilitate individuals’ entry into selfemployment.2 The design and implementation of such programmes is informed by numerous empirical studies that examine mostly the factors affecting individuals’ decision to become selfemployed, with relatively little information on the factors that determine entrepreneurial success and survival. High self-employment exit rates may be viewed as evidence of poor matches between entrepreneurial availability and skill requirements and can be costly, not only for the self-employed themselves but also for third parties, including banks, customers and governmental and private financial institutions. Understanding the determinants of self-employment survival could allow for a more complete response by policy makers who view entrepreneurship as the key to job creation and prosperity.

Electronic copy of this paper is available at: http://ssrn.com/abstract=941154 In this paper, we provide additional empirical evidence regarding the factors that determine self-employment duration in Britain, paying particular attention to non-pecuniary aspects of selfemployment and self-reported job satisfaction variables. Although recent empirical work on the determinants of self-employment entry acknowledges the importance of job satisfaction for individuals’ choice between salaried work and self-employment, the small number of studies examining self-employment survival focus exclusively on the role of standard economic and demographic variables. Generally, there is very little direct evidence on the correlation between self-reported satisfaction measures and self-employment duration. This is in sharp contrast to the empirical literature on labour turnover and mobility, where a growing number of studies examine the role of job satisfaction and non-pecuniary aspects of a job as strong predictors of quits [see for example, Akerlof et al (1988) and Clark et al (1998)]. More recently, Clark (2001), using employment spell data from the BHPS (omitting self-employment spells), finds that overall job satisfaction is indeed a powerful predictor of separations and quits, even after controlling for wages, demographic and other job variables. Interestingly, Clark (2001) also finds that job security and pay are the two job attributes most valued by workers.

Based on self-employment spell data from the BHPS, our results are generally consistent with the findings of previous studies as far as standard demographic and job variables are concerned. In particular, our results confirm the importance of assets and previous labour market status as important determinants of self-employment duration. Differences in survival rates across gender and different types of self-employment are also evident. However, we also find that job satisfaction and non-pecuniary characteristics of self-employment are strong predictors of selfemployment exit. Furthermore, when overall job satisfaction is broken down to five domain satisfaction variables, we find that satisfaction with pay, satisfaction with job security and satisfaction with work initiative emerge as the most important determinants of self-employment exit. This suggests that these three aspects of self-employment are the most valued by the selfemployed themselves. Similar findings emerge when estimating competing risk survival models, where exit probabilities to alternative destination states are compared. In general, the results of competing risks models lend support to the view that for many workers, self-employment is indeed a stepping stone to paid employment. Thus, the present paper contributes not only to increasing our general understanding of the determinants of self-employment survival, but also to a growing literature on the role of job satisfaction as a powerful predictor of quits and labour turnover, a rather novel aspect of the self-employment survival literature.

The paper is set out as follows: In section II, we describe the data and present some preliminary results based on non-parametric duration models. In section III, we present the results of single risk and competing risks proportional hazard models. In section IV, we conclude.

II. Data and Preliminary Results Our data are from Waves 1 to 8 (1991 to 1998) of the British Household Panel Survey (BHPS), a nationally representative sample of some 5,500 households and 10,000 individuals interviewed annually every fall since 1991. The BHPS is a rich data set providing information not only on a wide range of personal, household, and labour market characteristics, but also on lifetime job histories. Children in the households are interviewed separately once they reach the age of sixteen, and if any member of a household splits to form a new household, then all adult members of both the original and new households are interviewed annually. By such means, the survey remains broadly representative of the British populace over time.

We restrict our samples to include individuals who are between the ages of 18 and 64 for men and 18 and 60 for women. Individuals for whom information on the main variables of interest is not available are excluded from the sample (see Table 1: Note – all Figures and Tables are set out in Appendix B).

By combining information contained in the BHPS spell data files with employment status information collected at each wave, we construct a complete self-employment history for each individual in the sample. Our final sample consists of 1,436 self-employment spells that could be matched to individual interview data after cases with missing data and cases with inconsistencies have been removed.3 In this sample, 170 (11.83%) individuals have multiple spells and 1266 (88.17%) have a single spell for the period 1991-1998. Treating self-employment duration as a continuous variable, the data shows that the mean self-employment duration for males is 67.87 months (5.65 years) whereas the mean duration for females is 50.08 months (4.17 years). The corresponding median figures are 35 months (2.91 years) and 24 months (2 years) respectively.

Non-parametric survival analysis techniques allow us to examine the probability of an individual leaving the self-employment state conditional on having survived in the origin state for the whole interval. Figure 1 (shows the survivor function for men and women using the KaplanMeier estimator (Kaplan and Meier, 1958). The Kaplan-Meier estimates suggest that the selfemployed are more likely to quit during the first 100 months (8.3 years) and that females have lower survival rates than do males. The male hazard rate rises quickly above 400 months (33.3 years), probably due to retirement, while the female rate is somewhat flatter. Looking at the survivor functions, 51 percent (70 percent) of males (females) were eliminated from selfemployment within 200 months (16 years). Gender differences in survival rates may reflect gender differences in the motives for and aspirations from self-employment, differences in labour market opportunities, and even differences in tastes for pecuniary vs. non-pecuniary aspects of entrepreneurship. 4 Turning to the conditional probability of exiting to a new state, the survivor function, shown in Figure 2, illustrates how the exit probabilities vary with both time and the destination state under the competing risks assumption. We focus our attention to three main destinations states [notworking, including unemployment and family care (UNFC); wage employment (WE); and other self-employment (SE)], which account for 78.7 percent of all completed self-employment episodes.

In order to assess whether non-working is distinctly different from unemployment, we also examine the survivor function for transitions from self-employment to unemployment (UN). The probability of moving from self-employment to unemployment/family care (UNFC) is markedly higher than the probability of moving to any other destination states until the 144th month (12th year), where the probabilities of moving to unemployment/family care and wage employment coincide. The most likely destination state is wage employment, followed by unemployment/family care and then by self-employment, supporting the view that self-employment is a stepping stone to paid employment.

A possible explanation for this is that experience of self-employment unveils unobserved characteristics of the individual that increase the probability of finding a better match in the labour market.

Differences in the survivor functions between males and females are also confirmed by the Log-Rank test and Wilcoxon test for the equality of survivor functions (see Table 2).5 Both the LogRank test and the Wilcoxon test, reject the hypothesis of equality of the two survivor functions.

III. Econometric Analysis: Proportional Hazard Models In this section, we examine the determinants of self-employment duration by estimating Cox proportional hazard models [Cox (1972)]. The Cox proportional hazard model is semi-parametric, in the sense that it involves an unspecified baseline hazard instead of requiring further distributional assumptions. This approach, standard in applied work, assumes that censoring does not provide any information about latent failure times beyond that available in the covariates [e.g., Katz and Meyer (1990)]. The Cox proportional hazard model is also convenient for dealing with right censoring.

We model self-employment duration using both single risk and competing risks models. The former refers to the probability of exit from self-employment, irrespective of the state of destination. Competing risks models are used to estimate the probability of exit to different states, namely paid employment, unemployment/family care, unemployment and self-employment.6 Under the assumption of independent hazards, the competing risks model allows us to test whether the explanatory variables have different effects on the propensity to leave self-employment depending on the state of destination. All explanatory variables are measured at the last available interview prior to the termination of the spell.7 Results: Single Risk Proportional Hazard Model The results from estimating the single risk hazard model are shown in Table 3.8 As column 1 shows, males are more likely to survive in self-employment. Gender differences in self-employment survival could be attributed to differences in cultural attitudes towards entrepreneurship, risk-taking, and the availability and/or affordability of family services. Age has a significant and non-linear effect on self-employment survival. As Calvo and Wellisz (1980) suggest, age might be superior to labour market experience as both an indicator of the learning process and as an index of capital accumulation. Furthermore, Evans and Leighton (1989) claim that age may be a proxy for an individual’s attitude towards risk, with mature-aged persons being less willing to bear the stress and risks associated with self-employment. Taylor (1999) includes two age dummies (aged 30 and aged 30-49) and reports that male duration in self-employment is inversely related to age. Similar age effects are apparent for females, but the effect for both males and females disappears when the author desegregates his sample into voluntary and involuntary terminations. Consistent with the findings of Johansson (2001), our results show that there is a clear effect of age on the female risk of exiting self-employment, but such an effect is less pronounced for males.

Pages:   || 2 | 3 | 4 |

Similar works:


«A CFO’s Handbook on Performance Management transportation finance briefing papers providing up-to-date research results for state departments of transportation prepared for American Association of State Highway and Transportation Officials Center for Excellence in Project Finance prepared by Erik Cempel, Senior Associate Cambridge Systematics, Inc. American Association of State Highway and Transportation Officials 444 North Capitol Street, NW, Suite 249 Washington, DC 20001 (202) 624-5800...»

«Zeynep Bilgin Wührer’s Curriculum Vitae Lecturer at Institute for Marketing at Johannes Kepler University Linz / Austria zeynep.bilgin_wuehrer@jku.at; https://www.researchgate.net/profile/Zeynep_Bilgin Education Sep 1987 – Dec 1989 Hamburg UniversityHamburg, Germany DAAD Scholarship for PhD Thesis in International Marketing Sep 1985 – Oct 1990 Bogazici University Istanbul, Turkey PhD in Marketing, major and Organizational Behavior minor Sep 1983 – Jul 1985 Bogazici University Istanbul,...»

«African Journal of Accounting, Economics, Finance and Banking Research Vol. 7. No. 7. 2011. Lawrence Ogechukwu Obokoh, Chris Ehiobuche & A.E. Akinlo DICHOTOMY OF INTEREST RATE DEREGULATION POLICY IMPLICATIONS TO SME PERFORMANCE IN NIGERIA Lawrence Ogechukwu Obokoh1 University of Wales, UK. Email: Lechukwu2001@yahoo.com Chris Ehiobuche2 Berkeley College Garret Mountain Campus, USA. Email: cne@BerkeleyCollege.edu A.E. Akinlo3 Obafemi Awolowo University, Nigeria. Email: aakinlo@oauife.edu.ng,...»

«WP 2008-14 July 2008 Working Paper Department of Applied Economics and Management Cornell University, Ithaca, New York 14853-7801 USA Ethanol Plant Investment using Net Present Value and Real Options Analyses Todd M. Schmit, Jianchuan Luo, and Loren W. Tauer It is the Policy of Cornell University actively to support equality of educational and employment opportunity. No person shall be denied admission to any educational program or activity or be denied employment on the basis of any legally...»

«The People's Court Examined: A Legal and Empirical Analysis of the Small Claims Court System By BRUCE ZUCKER* AND MONICA HER** ONE OF THE most fascinating aspects of the legal system involves the small claims court process. Often referred to as the People's Court,' it is a part of the court that comes most directly into contact with the citizenry of a jurisdiction. Some cases are bizarre, most are interesting, and all are serious to the parties. 2 Provided that the dollar amount is less than or...»

«NBER WORKING PAPER SERIES WOMEN’S EMPOWERMENT AND ECONOMIC DEVELOPMENT Esther Duflo Working Paper 17702 http://www.nber.org/papers/w17702 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 December 2011 The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to...»

«Peace Processes and Statebuilding: Economic and Institutional Provisions of Peace Agreements This report was prepared for the World Bank and UNDP under the stewardship of the Chr. Michelsen Institute, Bergen (Norway) and reflects the independent views of the report's authors. Authors were Astri Suhrke, Torunn Wimpelmann and Marcia Dawes (UNDP consultant). The report represents an output of the UNDP and World Bank's joint programme on statebuilding (for 2006-7) and was discussed by the...»

«City of Berkeley Office of the City Auditor Ann-Marie Hogan, City Auditor Business License Tax Program Audit Report For the Fiscal Year Ending June 30, 2002 Prepared by: Ann Marie Hogan, City Auditor, CIA, CGAP Teresa Berkeley-Simmons, Audit Manager, CIA, CGAP Rosario Riche, Auditor I Gloria Ocampo, Auditor I Presented to Council May 13, 2003 2180 Milvia Street, Berkeley, CA 94704 Tel #(510) 981-6750 Fax (510) 981-6760 BUSINESS LICENSE TAX PROGRAM AUDIT REPORT For the Fiscal Year Ending June...»

«NORTH KIGEZI & KINKIIZI DIOCESES WATER AND SANITATION PROGRAMME (NKKD WATSAN) 1st Quarterly Report January March, 2015 Written by: NKKD WATSAN Team Section A: Basic Information The programme title North Kigezi & Kinkiizi Dioceses Water and Sanitation Programme (NKKD WATSAN) The improvement of health, alleviation of poverty and promotion of social and economic development in Rukungiri and Kanungu Districts of Southwest The programme mission Uganda through integrated, comprehensive and...»

«PRESS RELEASE #05 UPDATED CONSTRUCTION DOCUMENTATION VERSION 5.0 23 MAY 2012 Edited by Kata Dudás Supervised by Adrián Auth TABLE OF CONTENTS 3 ORGANIZATION CHART 4 LIST OF TEAM MEMBERS 5 Students 8 Professors 9 PROJECT DESCRIPTION 9 Use you forces 11 Team's organisation and objectives 13 Project development and current state 15 Dissemination activities and current impact 25 COLLABORATING INSTITUTIONS AND PARTNER COMPANIES PRESS RELEASE #05 | BUDAPEST UNIVERSITY OF TECHNOLOGY AND ECONOMICS...»

«    Opinions of the Supreme Court of Texas Fiscal Year 2014 (September 1, 2013 – August 31, 2014) Assembled by The Supreme Court of Texas Clerk’s Office Blake A. Hawthorne, Clerk of the Court Monica Zamarripa, Deputy Clerk FISCAL YEAR 2014 OPINIONS (SEPTEMBER 1, 2013 – AUGUST 31, 2014) 07-0010 GALVESTON CENTRAL APPRAISAL DIST. v. TRQ CAPTAIN'S LANDING 10-0121 THE FINANCE COMM'N OF TEX. v. NORWOOD 10-0567 TEXAS DEPT. OF HUMAN SERVICES v. OKOLI 10-0775 BOSTIC v. GEORGIA-PACIFIC CORP....»

<<  HOME   |    CONTACTS
2017 www.abstract.dislib.info - Abstracts, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.