جنبه های مالی و غیر مالی بقای خود اشتغالی

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.

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 self-employment. Based on data from the British Household Panel Survey (BHPS), Taylor (1999) finds that approximately 40% 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 self-employment.2 The design and implementation of such programmes is informed by numerous empirical studies that examine mostly the factors affecting individuals’ decision to become self-employed, 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. 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 self-employment 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, Rose, & Yellen, 1988; Clark, Georgellis, & Sanfey, 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 self-employment 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 self-employed 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 2, we describe the data and present some preliminary results based on non-parametric duration models. In Section 3, we present the results of single-risk and competing-risks proportional hazard models. In Section 4, we conclude.

A large literature in labour economics has examined the factors that affect individuals’ decision to become self-employed, and has informed the design and implementation of government programmes aiming to spur self-employment as the key to job creation and a way out of poverty. The success and efficiency of such programmes are best assessed by a rigorous analysis of self-employment duration. Unfortunately, however, relatively few empirical studies have examined the factors that affect self-employment survival and exit, and only one such study focuses on British self-employment (see Taylor, 1999). In this paper, we extend Taylor's (1999) work to provide additional empirical evidence on the determinants of self-employment duration in Britain. In our analysis, we have paid particular attention to non-pecuniary aspects of self-employment and self-reported job satisfaction variables as important determinants of self-employment survival. Apart from the role of standard economic and demographic variables, our results confirm the importance of job satisfaction variables in explaining self-employment survival probabilities for both males and females. Our results suggest that specific aspects of self-employment, such as job security, using initiative, and satisfaction with pay, are those most valued by the self-employed. It would therefore appear that a complete response by policy makers, who believe that self-employment could alleviate poverty and generate jobs, requires a closer look not only at the determinants of self-employment duration, but also a closer look at how the self-employed themselves evaluate their jobs.