Abstract
A growing concern exists with regard to the possibility that nonportability of employer-provided health insurance impedes self-employment and restricts business creation. In 1995, Taiwan implemented a National Health Insurance (NHI) program that extended health insurance coverage to all citizens. Such changes provide researchers with the opportunity to observe a natural experiment. Using a difference-in-differences regression on data from the Survey of Family Income and Expenditure in Taiwan, this paper examines the effects of universal health insurance on the likelihood of being an entrepreneur. We focus on two possible types of entrepreneurial activity: employers who hire workers and own-account workers. Results showed that implementation of NHI significantly decreased the incidence of own-account workers but increased the incidence of employers who hire workers. The best estimated possibility of being an employer increases by 3.3 percentage points, after NHI. Thus, the implementation of universal health insurance enables some new businesses, while inhibiting own-account workers. These findings should be informative for countries that plan to adopt a similar health policy.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Carree and Thurik (2010) provided an overview of the existing literature on the links between entrepreneurial activity and economic growth.
Most analysts agree that domestic political pressure is the main reason Taiwan was able to design and implement universal health reform in such a short time (Cheng 2003; Lu and Chiang 2011). The government lifted martial law in 1987 and Taiwan moved rapidly into democratization thereafter. To compete with the opposition party and gain popularity, the incumbent party (KMT) submitted its NHI bill to the Legislative Yuan in 1993, and the bill was passed in 1994.
GEI and FI each covered roughly 8% of the population prior to 1995. The government also provided fully subsidized health insurance to low-income households since 1990, but less than 1% of the population was covered. See Chou and Staiger (2001) for the comparison of these health insurance programs.
Entry to a civil service position in Taiwan required an individual to take an extremely competitive national exam. Those national exams were held once a year. The qualification rates (number of registered candidates, over number of qualified examinees) depended on the level and type of position. In general, the qualification rate was less than 10%.
Members of the famers’ association were required to be insured by FI, and farmers without the membership were still able to participate in the program under the examination of guidelines (Liao and Taylor 2010).
See Kan and Lin (2009) for details of the changes to premium rates before and after NHI.
The uninsured population at this time was mostly children, the elderly, and housewives.
With the exception of supplementary coverage for a few selected conditions such as cancer and accidents, no private health insurance was available (Chou et al. 2014). Health insurance is usually bundled with life insurance in Taiwan.
Taiwan’s NHI is characterized by high population coverage, comprehensive medical service coverage, and more importantly, very low cost. NHI has the lowest administration cost in the world of 2% (Wu et al. 2010).
For selecting a proxy for entrepreneurship, see Garcia (2014) for example.
To be consistent, we confine observations here to those analytic samples we utilize in empirical analysis later on. Please check Section 3 for sample selection and details. To further highlight the general trend, it is worth noting that we do not separate the sample by treatment and control group here.
Li et al. (2013) provided a theoretical model based on efficiency-wage theory and a compensation package of monetary wage and health insurance to explain why small firms are less likely to offer health insurance.
The sampling procedure was designed to represent the civilian noninstitutionalized population in Taiwan; 14,000–16,000 registered households were selected and interviewed during each study year.
Since the NHI represents a drastic change in healthcare policy, affecting all citizens, it will take time to exert its influence fully on the healthcare and labor markets. In order to evaluate the effects precisely, we may not choose the period to investigate carefully. Too short or long a period may skew results. The period 1995 to 2002 was deemed significant as the NHI rate increased from 4.25 to 4.55% in 2002. To be sure of significance, effects were estimated over various time periods and the results remained robust, despite that fact that “the NHI effect” is relatively moderate over shorter time periods. The results from different selections of time period are available from the author upon request.
House ownership is captured by the dummy variable. It is equal to 1 if the household owns a house and is 0 otherwise.
A probit model was also used to estimate the results and it was found that almost all the signs of coefficients are consistent with the LPM model. Compared with the probit model, the same statistical significance was obtained. The coefficients were easier to interpret when using the LPM; therefore, results are reported using those results. Probit results are available in Tables 9 and 10 in the Appendix.
In the baseline regression, we use several dummy variables to control for the group-specific heterogeneities in personal and family characteristics. However, the coefficients of these dummy variables are reflective of group means, which are not time-invariant. Thus, readers should interpret these coefficients with caution.
Yu (2004) used data from Manpower Utilization Surveys in Taiwan to analyze the decision to enter self-employment by male employees. Her findings showed that an increase in the unemployment rate pushed employed individuals toward self-employment in the 1980s, while the effects were not significant in the 1990s.
We also check the NHI effects by using the same empirical methodology for both being an employer and an own-account worker, stratified by different income levels, since NHI effects might be differentiable across different income groups. Despite the fact that the signs of the effect of NHI by the medium- and high-income level for being employer and own-account workers are consistent with our main results, the significance of coefficients is less clear. The confounding results could result from the fact that average annual family income is higher for the control group than for the treatment group. Thus, the NHI effects are contaminated due to sample selection problems.
We attempted “falsification” tests, to test the common trend assumption over the pretreatment period, which directly estimates any difference in trends between the treatment and control groups. Despite the fact that the results of the falsification test in some specifications show significantly different trends between the control and treatment group over the pretreatment period, the magnitude is minute, almost 0.
Stuart et al. (2014) discussed the conceptual issues associated with using propensity scores in difference-in-differences models to estimate the effects of a policy change. There are three main benefits of using the propensity score. First, using PSM can lead to more robust inferences (Ho et al. 2007). Second, the propensity scores condense the full set of covariates into a scalar summary, which makes the balancing approaches more feasible. Third, PSM is commonly used to reduce the potential bias in nonexperiment studies (Rosenbaum 2010; Rubin 2007).
The results indicate that the impact of NHI on entrepreneurial decision-making might be different among industrial sectors. The estimation of the NHI effect by industry is extended and discussed later in Section 4.2.3.
After we take the industry sector into consideration in column (4), the DID estimate became insignificant when compared to column (3). These results indicate that the impact of NHI on entrepreneurial decision-making may differ among industrial sectors. The estimation of the NHI effect by industry is extended and discussed later in Section 4.2.3. From Table 7, we see that the NHI effects by industry are insignificant in secondary and service sectors, but significant in the primary sector. However, the proportion available for observation in the primary sector is much lower than the proportion in secondary and service sectors (see Table 6). The insignificant DID effects in secondary and service sectors dominate the significant DID effect in the primary sector.
We also construct another treatment group by considering only nonworking wives to compare the effects of NHI with our origin treatment group—working wives in private sector. However, we do not find a distinct implication by considering only nonworking wives as another treatment group since the NHI effects are very similar to the original results. The results for comparing the NHI effects by different treatment groups are available from the authors upon request.
The reason why the DID effects in Table 4 column (4) became insignificant is very similar with Table 3. Similarly, we find that the NHI effects by different industries are also insignificant in secondary and service sectors, but only significant in primary sector in Table 8. However, the proportion of the observation in primary sector is also much lower than the proportion in secondary and service sectors (see Table 7). The insignificant DID effects in secondary and service sectors dominate the significant DID effect in the primary sector.
Hsieh et al. (2015) examined whether the introduction of NHI increases or decreases the likelihood of elderly parents residing with their adult children in Taiwan. They found that on average, having NHI reduced the probability of intergenerational co-residence, despite the fact that the effect may have been influenced by different types of socioeconomic family.
The variable, year, represents our selected sample periods and the details of the data set are mentioned in “the Data” section.
In our baseline regression model, we examine the determinants of self-employment by controlling various combinations of husbands’ and wives’ personal and family characteristics. To compare the estimates by different industries with the previous results, we use model (3) [results of column (3) in Tables 3, 4, and 5] as the basis of comparison including husbands’ and wives’ personal characteristics, such as age, education level, work area, and household control variables.
References
Abadie, A. (2005). Semiparametric difference-in-differences estimators. Review of Economic Studies, 72(1), 1–19. https://doi.org/10.1111/0034-6527.00321.
Angrist, J. D., & Pischke, J.-S. (2015). Mastering ‘metrics: the path from cause to effect. Princeton: Princeton University Press.
Baumol, W. J. (1990). Entrepreneurship: productive, unproductive and destructive. Journal of Political Economy, 98(5), 893–921. https://doi.org/10.1086/261712.
Becker, S. O., & Hvide, H. K. (2013). Do entrepreneurs matter? IZA Discussion paper no. 7146. https://doi.org/10.2139/ssrn.2199645.
Biehl, A. M., Gurley-Calvez, T., & Hill, B. (2014). Self-employment of older Americans: do recessions matter? Small Business Economics, 42(2), 297–309. https://doi.org/10.1007/s11187-013-9479-7.
Blundell, R., & Costa Dias, M. (2009). Alternative approaches to evaluation in empirical microeconometrics. Journal of Human Resources, 44(3), 565–640. https://doi.org/10.1353/jhr.2009.0009.
Blundell, R., Costa Dias, M., Meghir, C., & Van Reenen, J. (2004). Evaluating the employment impact of a mandatory job search program. Journal of the European Economic Association, 2(4), 569–606. https://doi.org/10.1162/1542476041423368.
Caliendo, M., Fossen, F., & Kritikos, A. S. (2014). Personality characteristics and the decisions to become and stay self-employed. Small Business Economics, 42(4), 787–814. https://doi.org/10.1007/s11187-013-9514-8.
Carree, M. A., & Thurik, A. R. (2010). The impact of entrepreneurship on economic growth. In D. B. Audretsch & Z. J. Acs (Eds.), Handbook of entrepreneurship 14 research (pp. 557–594). Berlin: Springer.
Cheng, T. M. (2003). Taiwan’s new national health insurance program: genesis and experience so far. Health Affairs, 22(3), 61–76. https://doi.org/10.1377/hlthaff.22.3.61.
Chou, Y. J., & Staiger, D. (2001). Health insurance and female labor supply in Taiwan. Journal of Health Economics, 20(2), 187–211. https://doi.org/10.1016/S0167-6296(00)00075-8.
Chou, S. Y., Grossman, M., & Liu, J. T. (2014). The impact of national health insurance on birth outcomes: a natural experiment in Taiwan. Journal of Development Economics, 111, 75–91. https://doi.org/10.1016/j.jdeveco.2014.08.004.
DeCicca, P. (2010). Health insurance availability and entrepreneurship. Upjohn Institute Working Paper No. 10-167. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. https://doi.org/10.2139/ssrn.1660435
Fairlie, R. W., Kapur, K., & Gates, S. W. (2011). Is employer-based health insurance a barrier to entrepreneurship? Journal of Health Economics, 30(1), 146–162. https://doi.org/10.1016/j.jhealeco.2010.09.003.
Fronstin, P., Helman, R., & Greenwald, M. (2003). Small employers and health benefits: findings from the 2002 small employer health benefits survey. Issue Brief No. 253.Washington, DC: Employee Benefit Research Institute (EBRI).
Garcia, A. B. (2014). Analyzing the determinants of entrepreneurship in European cities. Small Business Economics, 42(1), 77–98. https://doi.org/10.1007/s11187-012-9462-8.
Gilleskie, D. B., & Lutz, B. F. (1999). The impact of employer-provided health insurance on dynamic employment transitions. National Bureau of Economic Research.
Gruber, J., & Madrian, B. C. (1994). Health insurance and job mobility: the effects of public policy on job-lock. Industrial and Labor Relations Review, 48(1), 86–102. https://doi.org/10.2307/2524628.
Gruber, J., & Madrian, B. C. (1997). Employment separation and health insurance coverage. Journal of Public Economics, 66(3), 349–382. https://doi.org/10.1016/S0047-2727(96)01621-0.
Gruber, J., & Madrian, B. C. (2002). Health insurance, labor supply, and job mobility: a critical review of the literature. National Bureau of Economic Research.
Gumus, G., & Regan, T. (2007). Tax incentives as a solution to the uninsured: evidence from the self-employed. IZA Discussion paper no. 2866.
Heim, B. T., & Lurie, I. Z. (2009). Do increased premium subsidies affect how much health insurance is purchased? Evidence from the self-employed. Journal of Health Economics, 28(6), 1197–1210. https://doi.org/10.1016/j.jhealeco.2009.07.002.
Heim, B. T., & Lurie, I. Z. (2010). The effect of self-employed health insurance subsidies on self-employment. Journal of Public Economics, 94(11–12), 995–1007. https://doi.org/10.1016/j.jpubeco.2010.08.007.
Ho, D. E., King, I. K., & Stuart, E. A. (2007). Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political Analysis, 15, 199–236. https://doi.org/10.1093/pan/mpl013.
Holtz-Eakin, D., Penrod, J. R., & Rosen, H. S. (1996). Health insurance and the supply of entrepreneurs. Journal of Public Economics, 62(1–2), 209–235. https://doi.org/10.1016/0047-2727(96)01579-4.
Hsieh, H. L., Chou, S. Y., Liu, E., & Lien, H. M. (2015). Strengthening or weakening? The impact of universal health insurance on intergenerational coresidence in Taiwan. Demography, 52, 883–904. https://doi.org/10.1007/s13524-015-0387-0.
Huang, Y. H., & Hsueh, L. M. (1998). The impact of Taiwan’s national health insurance program on firms’ hiring and investment decisions. Taipei Economic Inquiry, 35(1), 1–28.
Ichino, A., Schwerdt, G., Winter-Ebmer, R. & Zweimuller, J. (2008). Too old to work, too young to retire? CEPR Discussion Paper No. DP6510.
Jackson, S. (2010). Mulling over Massachusetts: health insurance mandates and entrepreneurs. Entrepreneurship Theory and Practice, 34(5), 909–931. https://doi.org/10.1111/j.1540-6520.2009.00351.x.
Kan, K., & Lin, Y. L. (2009). The labor market effects of national health insurance: evidence from Taiwan. Journal of Population Economics, 22(2), 311–350. https://doi.org/10.1007/s00148-007-0135-x.
Lechner, M. (2010). The estimation of causal effects by difference-in-difference methods. Foundation and Trends in Econometrics, 4(3), 165–224. https://doi.org/10.1561/0800000014.
Li, C. S., Liu, C. C., Kuo, Y. C., & Yang, C. S. (2013). Health insurance provision and labor contracts for small firms. Small Business Economics, 40(2), 325–334. https://doi.org/10.1007/s11187-011-9371-2.
Liao, P., & Taylor, E. (2010). Health care reform and farm women’s off-farm labor force participation: evidence from Taiwan. Journal of Agricultural and Resource Economics, 35(2), 281–298.
Lu, J. F. R., & Chiang, T. L. (2011). Evolution of Taiwan’s health care system. Health Economics, Policy, and Law, 6(1), 85–107. https://doi.org/10.1017/S1744133109990351.
Madrian, B. C. (1994). Employment-based health insurance and job mobility: is there evidence of job lock? Quarterly Journal of Economics, 109(1), 27–54. https://doi.org/10.2307/2118427.
Morrisey, M., Jensen, G., & Morlock, R. (1994). Small employers and the health insurance market. Health Affairs, 13(5), 149–161. https://doi.org/10.1377/hlthaff.13.5.149.
National Federation of Independent Business. (2008). Small business problems and priorities. http://www.nfib.com.
Parker, S. C. (1996). A time series model of self-employment under uncertainty. Economica, 459–475. https://doi.org/10.2307/2555017.
Parker, S. C. (2004). The economics of self-employment and entrepreneurship. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511493430.
Rosenbaum, P. R. (2010). Design of observational studies. New York: Springer.
Rosenbaum, P. R., & Donald, B. R. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55. https://doi.org/10.1093/biomet/70.1.41.
Rubin, D. B. (2007). The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials. Statistics in Medicine, 26(1), 20–36. https://doi.org/10.1002/sim.2739.
Schumpeter, J. A. (1934). Theory of economic development: an inquiry into profits. Capital credit, interest and the business cycle. Cambridge: Harvard University Press.
Sheu, S. J., Chang, C. Y., & Kuo, Y. C. (2011). The impact of national health insurance on labor force participation of older men in Taiwan. Journal of Economics and Management, 7(2), 199–226.
Simoes, N., Crespo, N., & Moreira, S. B. (2016). Individual determinants of self-employment entry: what do we really know? Journal of Economic Surveys, 30(4), 783–806. https://doi.org/10.1111/joes.12111.
Stuart, E. A., Huskamp, H. A., Duckworth, K., Simmons, J., Song, Z., Chernew, M., & Barry, C. (2014). Using propensity scores in difference-in-differences models to estimate the effects of a policy change. Health Services & Outcomes Research Methodology, 14(4), 166–182. https://doi.org/10.1007/s10742-014-0123-z.
Velamuri, M. (2012). Taxes, health insurance and women’s self-employment. Contemporary Economic Policy, 30(2), 162–177. https://doi.org/10.1111/j.1465-7287.2011.00256.x.
Wellington, A. J. (2001). Health insurance coverage and entrepreneurship. Contemporary Economic Policy, 19(4), 465–478. https://doi.org/10.1093/cep/19.4.465.
Wu, T. Y., Majeed, A., & Kuo, K. N. (2010). An overview of healthcare system in Taiwan. London Journal of Primary Care, 3(2), 115–119. https://doi.org/10.1080/17571472.2010.11493315.
Yu, R. R. (2004). Transition to self-employment in Taiwan—a test of the recession-push hypothesis. Taiwan Economic Review, 32(1), 63–96. https://doi.org/10.6277/ter.2004.321.3.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Rights and permissions
About this article
Cite this article
Kuo, YC., Lin, JH. Picking the lock: how universal healthcare programs influence entrepreneurial activities. Small Bus Econ 54, 3–24 (2020). https://doi.org/10.1007/s11187-018-0077-6
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11187-018-0077-6