The Economics of Recession: A Survey
Part 6/10, October 2022 [1]
Arturo Estrella

6. Other systemic effects of recession

          Beyond the headline figures on real growth and unemployment, the effects of recessions can be pervasive throughout the whole economy. In this section, we take a look at research on systemic effects more broadly and consider possible impacts on industrial organization, financial investments, social relations, and public health. Perhaps surprisingly, not all the effects are detrimental, though some are difficult to pin down.

          Schumpeter (1942) proposed that recessions can have positive effects on industries by accelerating the process of elimination of weaker firms, for which he coined the term “creative destruction.”[2] Ouyang [R28] looks at two sides of this issue by allowing in his analysis for both a Schumpeterian “cleansing effect” as well as a “scarring effect” left by the destruction of potentially superior firms during their infancy. Calibrating the parameters of a theoretical model, the article concludes that the scarring effect dominates and that it reduces average productivity during recessions. At the very least, this analysis points to the limitations of relying exclusively on Schumpeter’s hypothesis to assess the costs of recession in the industrial sector.

          An aspect of the impact of recessions where formal research seems to be limited is the effect on financial investments. The bulk of the literature that deals with the relationship between financial asset prices and recessions has focused on the leading indicator properties of financial variables, which is discussed in the next section. That focus makes sense when we recognize that rational market participants need to look ahead when setting prices on financial instruments whose payments are scheduled to take place at various future horizons.

          The forward-looking nature of financial asset prices may also explain why it is difficult to identify consistent effects of recessions on these prices. For a simple illustration, consider the value of a single future dividend payment on an equity stake in an industrial firm. We can think of the present value of that payment as the product of the expected cash payment and a discount factor. During a recession, the expected value of the future cash payment may decline. However, the discount factor may have an offsetting higher value since interest rates are likely to fall in a recession. Moreover, the discount factor is also conceptually a function of a risk premium that may increase in the recession, in this case causing the discount factor to fall. The net result of these three effects leaves us with no a priori sense about the direction of the asset price movement in a recession. If we add to the mix the fact that expected cash flow, interest rates, and risk premiums each follows its own cyclical timing with phases that may not coincide, the asset price is even less likely to follow a consistent pattern through the recession.

          Some articles in the literature have tackled a limited aspect of the effect of recessions on financial asset prices. For instance, Estrella [R29] examines the cyclical pattern of optimal bank capital over the business cycle. In particular, the model looks at the effects of recessions on bank capital when capital is managed optimally and external capital is available to counteract asset losses. Using aggregate data for U.S. banks from 1984 to 2001, the article shows that banks raise more external capital during recessions, but that the new external capital is not enough to offset concurrent losses on bank assets. The net effect is a loss in the value of bank equity during the recession.

          Beber and Brandt [R30] investigate an indirect effect of recessions on bond prices. It is clear that prices of U.S. Treasury bonds may be affected by relevant news regardless of the stage of the business cycle. The strategy of this article is to look at expansions and recessions separately and to assess the effects of various types of macroeconomic announcements on bond returns in the two phases of the business cycle. From our perspective, the main result is that during recessions, good news about inflation seems to have the greatest impact on bond returns among the types of news tested.

          The next few readings consider a variety of social costs or benefits that may be associated with recessions. These effects do not impact income or employment directly but they affect society as a whole in ways that have the potential to change the level of aggregate welfare. For instance, theory suggests that price collusion by firms reduces the welfare of consumers vis-a-vis the case of perfect competition. Bagwell and Staiger [R31] construct a Markovian model of the business cycle in the style of Hamilton [R6] and investigate how incentives to collude change in the two phases. Their principal finding is that when market demand growth rates are positively correlated over time, the intensity of collusion is weakly procyclical in the sense that collusive prices are higher in expansions and lower in recessions. In this case, we can think of recessions as producing a mild social benefit.

          Barlevy [R32] takes issue with the Schumpeterian “cleansing effect” of recessions, as did Clark et al. [R24], but in this case the focus is on the quality of jobs. Through a phenomenon the article calls the “sullying effect” of recessions, the slow pace of creation of more productive jobs makes workers who hold the less productive jobs more likely to remain in place during a recession. They may still search for a better job, but it takes longer to find one in recession than in expansion. The end result does not necessarily lead to lower employment or loss of income directly, but it results in a greater proportion of lower quality, low productivity jobs.

          Household formation is an important determinant of housing demand in industrial societies. As offspring become more independent from their parents, there is a tendency for them to move on their own, with a resulting increase in housing demand. Lee and Painter [R33] construct a variety of alternative models to examine the connection between recessions and household formation. Based on calibration of the models using U.S. data, they find consistently across the models that recessions reduce the pace of household formation. The extent of the reduction is economically significant, with declines in household formation ranging from 1 to 9 percent, depending on the age of young adults during the recessionary period.

          Searing [R34] looks into the effects of recession on the more elusive social issue of interpersonal trust. Using data for Latin America from the World Values Survey, an international survey coordinated at the University of Michigan, Searing connects a quantitative measure of interpersonal trust to a set of control variables also available in the survey. Statistical regressions of the measure of trust on variables representing the stage and duration of the business cycle, as well as other control variables, suggest that interpersonal trust increases as recessions grow longer, resulting in a benefit rather than a cost. The regression variables control for confidence in the central government and church attendance, both of which are associated with higher levels of trust.

          A common perception in the late twentieth century was that individuals who grew up during the Great Depression tended to be cautious about personal finances and spending. Giuliano and Spilimbergo [R35] elaborate on this theme by investigating whether growing up in a recession has lasting effects on the economic and political views of individuals. Using 1972-2010 data from the U.S. General Social Survey, supplemented by the World Values Survey and a longitudinal survey of 1972 high school graduates, they indeed find statistical evidence that the effect of recessions on beliefs is long-lasting. In particular, they find that “individuals who experienced a recession when young believe that success in life depends more on luck than effort, support more government redistribution, and tend to vote for left-wing parties.”

          Disruptions caused by recessions can motivate some firms to enact practices intended to improve social well-being. Graddy-Reed and Feldman [R36] examine the responses of for-profit and non-profit enterprises in North Carolina to the 2008-09 recession in terms of social innovation. The latter is measured by alternative “scales of social investment” that quantify increases in the activity of the firms in various types of social practices falling into three general categories: environmental (11 practices), community (13 practices), and employee (13 practices). Results indicate that the “more socially innovative organizations,” which have a history of activity in the social practices, tended to introduce further social innovation in all three categories in response to the 2008-09 recession.

          The next four readings address the effect of recessions on various health-related issues. Ruhm [R37] goes to the heart of the matter by asking “Are Recessions Good for Your Health?” The simple, perhaps surprising answer is that yes, they seem to be. Using a longitudinal survey from 1972 to 1991 as well as behavioral microdata from 1987 to 1995 for the United States, the article concludes that total mortality, as well as fatalities from 8 of 10 causes examined, tend to fall during recessions. Moreover, the effects are greater for causes and age groups for which behavioral reasons are most plausible, such as motor vehicle fatalities and young adults. One caveat to these results is that long economic expansions, though not shorter ones, are also associated with decreased mortality rates. A second caveat is that Ruhm (2015) finds in follow-up work that the pattern in recessions is reversed for some causes, including cancer, though it remains the same for cardiovascular disease and transport accidents.[3]

          Boone and van Ours [R38] consider a stylized fact that workplace safety seems to improve during recessions and find conflicting evidence by peering below the surface. Their analysis, using data for 16 OECD member countries, suggests that the rate of workplace accidents does not decrease during recessions, but that it is rather the rate at which these accidents are reported that declines. They surmise that reporting an accident has a negative effect on the reputation of a worker and increases the probability that the worker will be fired, in both cases reducing the incentive to report. The effect is particularly strong during recessions, when job security is more vulnerable.

          NcInerney and Mellor [R39] return to the question of whether recessions affect mortality rates, but this time focusing the senior population, aged 65 and above. Their main source of information is the U.S. Medicare Beneficiary Survey for the period from 1994 to 2008. In contrast to the earlier literature (Cf., [R37]), they find that mortality increases during recessions among the senior population. In addition, seniors tend to report worse mental health during recessions and there is no tendency to engage in healthier behaviors.

          Do declines in disposable income during recessions deter individuals from engaging in costly unhealthy behavior? Using longitudinal data for Canada from 1994 to 2009, Latif [R40] finds that quite the contrary applies to alcohol consumption and smoking during recessions. Results show that a higher unemployment rate has a significant positive impact on alcohol consumption as well as on the likelihood of becoming a binge drinker. Cigarette consumption among daily smokers increases as well, though in this case the higher unemployment rate does not increase the likelihood of becoming a smoker.

          On balance, is there a silver lining to recessions from these other systemic effects? Many of them are negative, reinforcing declines in output and employment. The encouraging exceptions are collusive pricing ([R31]), interpersonal trust ([R34]), social innovation ([R36]), and overall public health ([R37]). Even in these cases, however, we have seen that there are caveats to consider.

Readings referenced from book The Economics of Recession

R28    Min Ouyang (2009), ‘The Scarring Effects of Recessions’, Journal of Monetary Economics, 56 (2), March, 184–99

R29    Arturo Estrella (2004), ‘The Cyclical Behavior of Optimal Bank Capital’, Journal of Banking and Finance, 28 (6), June, 1469–98

R30    Alessandro Beber and Michael W. Brandt (2010), ‘When it Cannot Get Better or Worse: The Asymmetric Impact of Good and Bad News on Bond Returns in Expansions and Recessions’, Review of Finance, 14 (1), January, 119–55

R31    Kyle Bagwell and Robert W. Staiger (1997), ‘Collusion over the Business Cycle’, RAND Journal of Economics, 28 (1), Spring, 82–106

R32    Gadi Barlevy (2002), ‘The Sullying Effect of Recessions’, Review of Economic Studies, 69 (1), January, 65–96

R33    Kwan Ok Lee and Gary Painter (2013), ‘What Happens to Household Formation in a Recession’, Journal of Urban Economics, 76, July, 93–109

R34    Elizabeth A.M. Searing (2013), ‘Love Thy Neighbour? Recessions and Interpersonal Trust in Latin America’, Journal of Economic Behavior and Organization, 94, October, 68–79

R35    Paola Giuliano and Antonio Spilimbergo (2014), ‘Growing Up in a Recession’, Review of Economic Studies, 81 (2), April, 787–817

R36    Alexandra Graddy-Reed and Maryann P. Feldman (2015), ‘Stepping Up: An Empirical Analysis of the Role of Social Innovation in Response to an Economic Recession’, Cambridge Journal of Regions, Economy and Society, 8 (2), July, 293–312

R37    Christopher J. Ruhm (2000), ‘Are Recessions Good for Your Health?’, Quarterly Journal of Economics, 115 (2), May, 617–50

R38    Jan Boone and Jan C. van Ours (2006), ‘Are Recessions Good for Workplace Safety?’, Journal of Health Economics, 25 (6), November, 1069–93

R39    Melissa McInerney and Jennifer M. Mellor (2012), ‘Recessions and Seniors’ Health, Health Behaviors, and Healthcare Use: Analysis of the Medicare Current Beneficiary Survey’, Journal of Health Economics, 31 (5), September, 744–51

R40    Ehsan Latif (2014), ‘The Impact of Recession on Drinking and Smoking Behaviours in Canada’, Economic Modelling, 42, October, 43–56