AP Images

About the authors: Kyle Herkenhoff is an assistant professor of economics at the University of Minnesota. Gordon Phillips is the C.V. Starr Foundation Professor and Academic Director of the Center for Private Equity and Entrepreneurship at the Tuck School of Business, Dartmouth College.

The aim of this essay is to provide several pertinent facts about the way unemployed households in the United States use consumer credit and the way bankruptcy flags affect job finding rates and business creation. These facts can be used by policymakers, including legislators and central bankers, in order to better understand the implications and feasibility of both consumer credit regulations and monetary policy.

The basis for these facts is a new dataset whose construction was funded by the National Science Foundation, implemented in large part by one of the co-authors of this essay, Gordon Phillips, and University of Maryland finance professor Ethan Cohen-Cole, and recently analyzed in joint work with the other co-author of this issues brief, Kyle Herkenhoff.1 There were four major observations and implications that came out of our dataset:

  1. Credit cards are a form of unemployment insurance
  2. Expansionary monetary policy (lowering interest rates) may give unemployed consumers more ‘breathing room’ and allows them to find jobs at higher paying, larger, and more productive firms
  3. Access to consumer credit facilitates self-employment as well as the transition into hiring an entrepreneur’s first employee
  4. Bankruptcy flags disrupt job finding, business creation, and reallocation of workers across jobs

In the remainder of this essay, we explain each of these findings, the circumstances under which they obtained, and the implications for policymakers and lawmakers.

Credit cards are a form
of unemployment insurance

Our first main finding is that consumer credit (credit cards, personal revolving loans, and other forms of revolving credit) has an effect on unemployed households that is comparable to unemployment insurance. In simple terms, being able to borrow allows unemployed households to search more thoroughly for a job. Just like unemployment insurance, credit cards and other forms of revolving credit allow unemployed individuals to “hold themselves over” by, say, buying groceries, or in economic terminology, it allows them to “smooth consumption.” Therefore, consumer credit allows them to find better job matches, and, as a consequence, they are paid higher wages.

We begin with a sample of 3 million workers. We first focus on a set of 20,000 displaced workers, some of whom have significant amounts of credit limits, while others have very limited consumer credit access. We use exogenous increases in credit that result from the removal of bankruptcy flags and from automatic increases in credit to isolate credit increases that are not related to an individual’s job prospects and their underlying employability or quality. We find that the more credit unemployed workers have, the longer they take to find a job. Among those who find a job, they find jobs with higher wages.

Delivering equitable growth

Previous article: Supply chains, Susan Helper
Next article: Wealth transfer taxation, Lily Batchelder

These findings suggest that consumer credit acts in a very similar way to unemployment insurance. Existing unemployment insurance studies find that unemployment insurance protracts unemployment durations, and workers generally find higher paying jobs. (This is true in the United States and Europe;2 European estimates, however, are sometimes insignificant or negative).3 The similarity between the way unemployed individuals use consumer credit and unemployment insurance suggests that households have some degree of private insurance against job loss through credit markets, and that government programs in which consumer credit is extended to unemployed individuals rather than as a transfer payment may produce similar disincentive effects.

Expansionary monetary policy may give unemployed
consumers more breathing room to find better jobs

What are the implications of our findings for monetary policy? The new dataset allows us to measure the impact of consumer credit access on labor market outcomes for the first time. Our results suggest that if interest rates are lowered, or if the government provides some more “breathing room” for unemployed consumers, then they may take longer to find a job, and they may ultimately find better job matches. A direct consequence of this mechanism is that if the government lowers interest rates then the unemployment rate may initially increase. With lower interest rates, and a greater ability to smooth consumption, households may be able to hold themselves over while searching more thoroughly for a job.

Consequently, the duration of unemployment and the unemployment rate will initially be higher following an interest rate decline. Yet the wages of those workers who find jobs will be higher because they are searching more thoroughly and finding better matches. Thus, our research suggests that a central economic-performance indicator of the Federal Reserve should be the wages of new hires, not necessarily the unemployment rate. We believe that by focusing on measures of match quality, the Federal Reserve can take into account the role that credit plays in household job-search decisions as well as have a more complete picture of the health of new labor market matches.

Access to consumer credit facilitates self-employment as
well as the transition into hiring an entrepreneur’s first employee

Consumer credit is not just used to facilitate a thorough job search, it is also a critical component of financing for the self-employed and job creation.4 To examine the importance of consumer credit for the self-employed, we build another new dataset with 200,000 individuals who have previously filed for bankruptcy and link these individuals to Internal Revenue Service entrepreneur tax records with administrative employment histories, credit histories, and so called SS-4 IRS business ownership data.

Using this dataset, we are able to follow individuals over time and observe all possible employment transitions, comprised of: transitions in and out of working for another business; transitions in and out of self-employment; and transitions in and out of owning an employer firm in the Integrated Longitudinal Business Database, or ILDB, which is the merged dataset of SS-4 ownership records with firm employment.5 Our main source of exogenous variation in credit access comes from the removal of consumers’ bankruptcy flag.6 We show that following bankruptcy flag removal, individuals receive a large increase in consumer credit access. Following this large, discrete, and unanticipated increase in consumer credit access,7 we find what we call the “credit access effect.”

Bankruptcy flags disrupt job finding, business
creation, and reallocation of workers across jobs

We show, for the first time to our knowledge, that consumer credit is critical for making the leap from a non-employer business to an employer business. In other words, consumer credit facilitates the hiring of the first initial employee, allowing individuals to make the transition out of self-employment into becoming a job-creating entrepreneur. Specifically, we find that:

  • Flows into self-employment increase disproportionately after credit access improves. Those individuals who start new businesses earn 4 percent more Schedule C Net Income ($1,000) versus comparable bankrupt individuals who have not yet had their bankruptcy flag removed.
  • Following the discrete rise in credit access, these individuals are more likely to become employer firms in the Integrated Longitudinal Business Database
  • New-firm owners in this database borrow $40,000 more using mortgages and home equity lines of credit

These findings suggest that consumer credit matters for the subgroup of individuals who want to start a business, and moreover, it matters disproportionately for those individuals who want to grow their businesses.

A crucial fact for subsequent mortgage regulation is that self-employed individuals who make their initial hire borrow $40,000 more than the control group, and they primarily use mortgage credit to facilitate this transition. In particular, they borrow using home equity lines of credit and other forms of high-interest-rate revolving credit.

This is an important set of facts for regulatory institutions, such as the Consumer Finance Protection Bureau, because this implies that restrictions on access to mortgage credit have direct implications not just for “mom-and-pop” self-employed individuals but also for those who intend to grow rapidly and hire additional employees. Consumer credit may not be the only source of financing for these businesses, but our results indicate that it is, on average, an important part of the debt portfolio of young, growing firms.

The United States is currently witnessing a long-run trend decline in startups.8 By further curtailing or restricting consumer credit, startup rates (and in particular high-growth startup rates) may drop. Our research therefore calls for follow-up studies on regulations that the CFPB may consider, and in particular, mortgage restrictions, especially home equity lines of credit.

Using the same dataset, we are able to measure the impact of bankruptcy flag removal on employment prospects and wages as well as on self-employment and business income. Our final policy relevant finding on the topic of consumer credit is that bankruptcy flags are likely misallocating workers across sectors. Using the same dataset of 200,000 individuals who previously filed for bankruptcy, we are able to study the way bankruptcy flag removal affects labor markets, self-employment, and earnings. We notice four broad patterns following bankruptcy flag removal:

  • Individuals flow into formal sector unemployment-insured jobs. In simple terms, following bankruptcy flag removal, individuals find jobs that qualify them for unemployment insurance. These jobs provide a safety net to the worker in the case of job loss.
  • Those who flow into formal-sector jobs after bankruptcy flag removal earn significantly more and are extremely attached to the formal sector. In simple terms, they earn more and are less likely to end up non-employed than are other comparable individuals without a flag removal.
  • Individuals flow out of “informal” sector self-employed jobs. In simple terms, individuals leave self-employment after bankruptcy flag removal, and they subsequently find jobs in the formal sector.
  • Individuals also flow into “informal” sector self-employed jobs (as mentioned above). With greater credit access, nascent entrepreneurs can quit their formal sector jobs and use credit to finance their ideas.

The main policy implication of our bankruptcy-flag removal findings has to do with the current debate over the use of credit checks by human resource departments.9 Our results indicate that after bankruptcy flag removal, there is a significant amount of reshuffling of workers across sectors. In economic terms, there appears to be reallocation, although whether this is a welfare-improving reallocation remains to be determined. Based on the wages of new hires and their subsequent job transitions (especially the fact that they do not exit to non-employment), our findings suggest workers with bankruptcy flags are not going to the jobs that value them the most. We therefore suggest to policymakers who are considering credit-check bans to consider the impediments that bankruptcy flags generate for self-employment, formal-employment, and new employment in their cost-benefit analyses.

  1. Kyle F Herkenhoff, Gordon Phillips, and Ethan Cohen-Cole, “How credit constraints impact job finding rates, sorting & aggregate output,” Manuscript, 2015. See also Kyle F Herkenhoff, Gordon Phillips, and Ethan Cohen-Cole, “The impact of consumer credit access on employment, earnings and entrepreneurship,” Manuscript, 2016.
  2. Raj Chetty, “Moral hazard vs. liquidity and optimal unemployment insurance,” Working Paper, (Cambridge, MA: National Bureau of Economic Research, 2008). See also Arash Nekoei and Andrea Weber, “Does extending unemployment benefits improve job quality?” Working Paper, (Social Science Research Network, 2015).
  3. Johannes Schmieder, Till Von Wachter, and Stefan Bender, “The causal effect of unemployment duration on wages: Evidence from unemployment insurance extensions,” Working Paper, (Cambridge, MA: National Bureau of Economic Research, 2013).
  4. Allen N Berger and Gregory F Udell, “The economics of small business finance: The roles of private equity and debt markets in the financial growth cycle,” Journal of Banking & Finance, 22, no. 6 (1998): 613–673. See also Robert W Fairlie and Harry A Krashinsky, “Liquidity constraints, household wealth, and entrepreneurship revisited,” Review of Income and Wealth, 58, no.2 (2012): 279–306; M. Adelino, K. Gerardi, and P.S. Willen,” “Why don’t lenders renegotiate more home mortgages? redefaults, self-cures and securitization,” Working Paper, (Cambridge, MA: National Bureau of Economic Research, 2009); and Alicia M Robb and David T Robinson, “The capital structure decisions of new firms,” Review of Financial Studies 27, no. 1 (2014).
  5. Steven J Davis, John Haltiwanger, Ron S Jarmin, Cornell J Krizan, Javier Miranda, Alfred Nucci, and Kristin Sandusky, “Measuring the dynamics of young and small businesses: Integrating the employer and nonemployer universes,” (Working Paper: National Bureau of Economic Research, 2007).
  6. D.K. Musto,” What happens when information leaves a market? evidence from post bankruptcy consumers,” The Journal of Business, 77, no. 4 (2004): 725–748.
  7. Tal Gross, Matthew J. Notowidigdo, and Jialan Wang, “The marginal propensity to consume over the business cycle,” Manuscript, 2016.
  8. Ryan Decker, John Haltiwanger, Ron Jarmin, and Javier Miranda,” The secular decline in business dynamism in the United States,” Manuscript, (College Park, MD: University of Maryland, 2013). See also Fatih Karahan, Benjamin Pugsley, and Aysegül Sahin,” Understanding the 30-year decline in the startup rate: A general equilibrium approach,” Unpublished manuscript, May, 2015.
  9. Daphne Chen, Dean Corbae, and Andy Glover, “Can employer credit checks create poverty traps?” Unpublished Manuscript, 2013. See also Daniel Shoag and Robert Clifford, “‘No more credit score’: Employer credit check bans and signal substitution,” Working Paper, (Social Science Research Network, 2016); Marieke Bos, Emily Breza, and Andres Liberman,” The labor market effects of credit market information,” Working Paper, (Social Science Research Network, 2015); Kristle Cortes, Andrew Glover, and Murat Tasci, “The unintended consequences of employer credit check bans on labor and credit markets,” Manuscript, 2016; and Will Dobbie, Paul Goldsmith-Pinkham, Neale Mahoney, and Jae Song, “Bad credit, no problem? credit and labor market consequences of bad credit reports,” Working Paper, (Social Science Research Network, 2016).