Wednesday, June 4, 2008

The Great Moderation vs. Rising Household Income Volatility

Just last week I revisited the question of how to reconcile the findings from the "Great Moderation" literature that shows a significant decline in aggregate economic volatility since the early-to-mid 1980s with the findings of Jacob Hacker and others that show there has been a marked increase in household income volatility over this same period. One would think some of the decreased macroeconomic volatility would be experienced and observed at the household level. The data, however, says otherwise. How is this possible? A new paper on the "Great Moderation" by Steven Davis and James Kahn attempts, among other things, to answer this question.

From their paper, Interpreting the Great Moderation: Changes in the Volatility of Economic Activity at the Macro and Micro Levels, we get the following discussion of this issue:
[A] puzzle that research on the Great Moderation has yet to confront: Why has the dramatic decline in the volatility of aggregate real activity, and the roughly coincident decline in firm-level volatility and job-loss rates, not translated into sizable reductions in earnings uncertainty and consumption volatility facing individuals and households?

We do not know the answer to this question, but we conjecture that greater flexibility in pay setting for workers played a role, possibly a major one. Greater pay flexibility is consistent with the rise in wage and earnings inequality in U.S. labor markets since 1980 and with increases in individual income volatility and earnings uncertainty. If these developments involve a rise in the variance of idiosyncratic permanent income shocks to households, then household consumption volatility also rises according to permanent income theory. Greater wage (and hours) flexibility also leads to smaller firm-level employment responses to idiosyncratic shocks and smaller aggregate responses to common shocks, because firms can respond by adjusting compensation rather than relying entirely on layoffs and hires. By the same logic, wage adjustments can substitute for unwanted job loss. So, at least in principle, greater wage flexibility offers a unified explanation for the rise in wage and earnings inequality, flat or rising volatility in household consumption, a decline in job-loss rates, and declines in firm-level and aggregate volatility measures.
In short, their argument is that greater wage volatility has been traded for reduced output and employment volatility. If true, this interpretation has two implications: (1) labor markets are working better since the price of labor is now more flexible; (2) more economic risk has been shifted to labor.

Read the rest of the article.

Update I: See further discussion of this posting here.


  1. Thanks for this interesting post.
    The increase in consumption volatility suggests that the increase in the variance of wages is due to the permanent component rather than to the transitory one.
    In a more flexible wage world, the wage should closely mimic productivity. I suspect that wage shocks are more persistent than productivity shocks.
    For the increased efficiency in the labor market to be a resolution of the puzzle, shouldn’t the wage volatility be less persistent, mimicking transitory productivity shocks, and therefore having negligible effects on consumption?
    Unless it is possible to identify firm level productivity shocks as permanent as the wage shocks, this argument would go against the resolution of the puzzle proposed by Davis and Kahn.
    The wage flexibility wouldn’t in fact explain the observed increase in the wage volatility.

  2. Alex,

    Thanks for your insights.

    Mark Thoma put this post over at his blog and it received a lot of discussion. I will put up a link to this discussion at his site.

  3. We use a dynamic factor model estimated on quarterly state-level data from 1986 to 2005 via Bayesian methods to disentangle the relative importance of the common component in OFHEO house price movements from local (state- or region-specific) shocks. We find that historically movements in house prices were mainly driven by the local component. The recent period (2001–2005) has been different: the increase in house prices is a national phenomenon. We use a VAR to investigate the extent to which expansionary monetary policy is responsible for this phenomenon. We find the impact of policy shocks on house prices to be small in comparison with the magnitude of recent fluctuations.

    house for sale by owner