Linking Wage Labor and Savings: Evidence from Sri Lanka

We often talk about how access to financial instruments may complement entrepreneurship.  Financial instruments such as vehicles for savings and loans may help to encourage entrepreneurship and investment by making it possible for individuals to make larger investments and to hoard returns for the future.  Less has been said about the interaction between financial access and wage work, but a recent paper by Michael Callen, Suresh De Mel, Craig McIntosh and Christopher Woodruff shows, perhaps surprisingly, that a strong link can exist between financial access and wage labor as well.

In their experimental study, individuals in Sri Lanka were offered access to an improved savings product in which weekly deposits could be made to deposit collectors operating door-to-door with digital point-of-service terminals to record deposits.  As in previous studies, access to the savings product increases savings and expenditures.  The authors however also find that access to this savings product increased incomes while simultaneously encouraging disinvestment in microenterprises.  Individuals with access to the savings product, relative to the comparison group, were more likely to engage in wage work, increasing incomes while decreasing micro-entrepreneurship.

In the paper, incomes rise by approximately 10 percent relative to baseline while labor hours in wage labor rise 20 percent.  The authors propose a simple theoretical framework within which an increase in returns to saving may generate both rising work hours and a shift from self-employment to wage work.  If there are constant returns to effort in the wage sector but diminishing marginal returns to effort in the entrepreneurial sector, then an increase in the effective interest rate on savings will induce people to shift their equilibrium allocation of labor from entrepreneurship to wage labor.  In this paper, the authors conjecture that it is these adjustments in labor allocation and related increases in income that account for the substantial increases in savings associated with access to savings product and service.

Overall savings increase by approximately one-third of the increase in incomes, a remarkable effect.  The authors posit that increases in labor income are in fact the source of increases in savings in their data, as consumption and transfers to others do not appear to change with access to the point-of-service savings accounts.  In contrast, access to the point-of-service linked savings accounts was associated with a significant increase in monthly consumption.

The paper carefully documents the effects of access to formal savings through these point-of-service savings accounts on participation in and contributions to informal savings groups, and finds no adverse effects of access to formal savings on these outcomes.  The authors randomized the density of treatment with the new savings accounts within savings groups and found that participation in savings groups increases with access to the new savings accounts and that savings through savings groups also increases.  This suggests a complementarity between formal and informal savings mechanisms that has not been shown or highlighted in earlier papers.  In contrast, Duflo and Robinson (2013) find evidence for long-run substitutability between individual and group savings mechanisms.  This paper instead provides optimistic evidence that individuals may take advantage of both to better their savings outcomes.

Overall, the paper provides a tantalizing hint that the effects of financial access may be broader than we have been prepared to think, affecting labor market choices in a way that may be more fundamental than simple increases in investment in the status quo business for micro- or small-scale entrepreneurs.  These effects may be context dependent, for example having more of a presence where wage labor markets are thick, but they are important to document as financial access may have a broader set of effects than we have traditionally considered.  It shows that access to formal savings products need not crowd out informal savings and it also raises the important policy point that the current focus on the nexus of financial access and entrepreneurship may miss some important ways in which financial access can benefit the poor.  We look forward to seeing more work on employment and labor supply decisions in financial access papers to come.