Dissertation. For this, I get to call myself "Doctor", and I don't even have to learn to play golf.
Has There Been a Great Risk Shift? Trends in Economic Instability Among Working Age Adults
(the full kit and kaboodle -- the answer, in case you're wondering, is No)
Chapter 1 - Introduction
Discusses sources of economic risk and instability and reviews additional considerations in thinking about economic risk (availability of insurance against risk, anticipated vs. unanticipated instability, living standards vs. instability, and moral hazard in attempting to insure against risk). Measurement of economic instability.
Chapter 2 - It Ain't Got That Swing: Trends in Earnings Instability and Volatility
For the typical male, earnings volatility probably increased by no more than a third between the early 1970s and the early 2000s. Among men without self-employment earnings, the increase in volatility was confined to the 1970s and early 1980s. The risk of a sudden drop in earnings grew modestly through the early 1980s but has been flat since. Among women, labor income volatility declined by as much as a quarter. The timing of the increase among men does not match accounts that claim wage and salary workers' security has recently deteriorated. Nor does the overall increase appear as large as some previous studies have implied. The typical earnings reversal over a short-term window increased only from 12.5 percent to 13.8 percent, and even when self-employment earnings are included, the increase was only from 14.1 to 17.7 percent. Other evidence suggests that volatility levels were probably higher prior to 1960 than after 1970, and they were probably similar in 2008 as in 2002 or 2004.
Chapter 3 - The Very Small Risk Shift: Trends in Family Income Changes
I first discuss my (uncredited) discovery of the flaws in Jacob Hacker's influential work. I then turn to my own analyses. I find very little evidence for the hypothesis that there has been a sizable shift for the worse in the risk of negative family income shocks or volatility faced by working-age adults. Consistent with findings based on administrative data, the risk of an income drop increased little if at all from the early 1970s to the early 2000s, and volatility rose only modestly, by 10 to 35 percent. Other evidence on trends in economic risk suggests that income instability and volatility was higher prior to 1960 and no higher in 2008 than in 2002 or 2004. Hacker's original claim was that volatility had risen 200 percent over thirty years, later revised downward to 100 percent. I find that the typical family's income reversals—up over two years then down, or vice versa—within a nine-year window have increased from 15 or 16 percent of income, up and down, to 17 or 18 percent.
Chapter 4 - Shaky Ground? Trends in the Risk of Large Earnings and Income Drops Among Demographic Groups
I examine trends in the risk of large earnings and income drops for several different demographic groups in order to determine the extent to which trends and levels of instability are accounted for by those variables. The results indicate that instability is highest among less-educated men and women, black men, new mothers, and women whose relationship ends. Over time, instability rose among men within each educational category, among less educated women, among men experiencing a breakup, and among families headed by an adult under age 40 or a college graduate. Differences in trends across demographic groups explain very little of the overall trends in instability observed in Chapters Two and Three. Rather, higher educational attainment and, among women, declining and delayed fertility and relationship formation shifted Americans into lower-instability demographics over time.
Chapter 5 - Conclusion: Economic Risk, Policy, and Politics
Trends in other indicators of risk (joblessness, health care coverage and affordability, retirement security, and indebtedness) confirm that economic risk has not grown substantially in recent decades. Economic insecurity has grown since the early 1990s recession, but evidence suggests that this may be due to disproportionate coverage of downsizing in the media. Levels of economic insecurity are comparable to levels of anxiety about other risks such as dying in a plane crash or becoming a victim of crime, and people generally say their own economic situation is much better than their assessment of others' situations. This chapter also illustrates flaws in Jacob Hacker's and Paul Pierson's Off Center, showing that the range of policies Americans will support is wide enough to incorporate (most of) the domestic agenda of George W. Bush and the GOP Congress of the 2000s. It is also wide enough to incorporate left-leaning policies, though Hacker and Pierson (and others) overstate the strength of support for such policies. Finally, I propose a system of "citizen benefits" designed to address economic insecurity while attracting widespread support and not blowing up federal deficits any further.
Appendix 1 - Previous Research on Earnings Instability and Volatility Trends
Appendix 2 - Previous Research on Income Instability and Volatility Trends
Files for Replication
The zip archive contains a number of STATA do files. Start by reading "master.do", which explains how to obtain my PSID extracts and other data, indicates which do files apply to which chapters, and explains how to set up the necessary files to obtain results.
(the full kit and kaboodle -- the answer, in case you're wondering, is No)
Chapter 1 - Introduction
Discusses sources of economic risk and instability and reviews additional considerations in thinking about economic risk (availability of insurance against risk, anticipated vs. unanticipated instability, living standards vs. instability, and moral hazard in attempting to insure against risk). Measurement of economic instability.
Chapter 2 - It Ain't Got That Swing: Trends in Earnings Instability and Volatility
For the typical male, earnings volatility probably increased by no more than a third between the early 1970s and the early 2000s. Among men without self-employment earnings, the increase in volatility was confined to the 1970s and early 1980s. The risk of a sudden drop in earnings grew modestly through the early 1980s but has been flat since. Among women, labor income volatility declined by as much as a quarter. The timing of the increase among men does not match accounts that claim wage and salary workers' security has recently deteriorated. Nor does the overall increase appear as large as some previous studies have implied. The typical earnings reversal over a short-term window increased only from 12.5 percent to 13.8 percent, and even when self-employment earnings are included, the increase was only from 14.1 to 17.7 percent. Other evidence suggests that volatility levels were probably higher prior to 1960 than after 1970, and they were probably similar in 2008 as in 2002 or 2004.
Chapter 3 - The Very Small Risk Shift: Trends in Family Income Changes
I first discuss my (uncredited) discovery of the flaws in Jacob Hacker's influential work. I then turn to my own analyses. I find very little evidence for the hypothesis that there has been a sizable shift for the worse in the risk of negative family income shocks or volatility faced by working-age adults. Consistent with findings based on administrative data, the risk of an income drop increased little if at all from the early 1970s to the early 2000s, and volatility rose only modestly, by 10 to 35 percent. Other evidence on trends in economic risk suggests that income instability and volatility was higher prior to 1960 and no higher in 2008 than in 2002 or 2004. Hacker's original claim was that volatility had risen 200 percent over thirty years, later revised downward to 100 percent. I find that the typical family's income reversals—up over two years then down, or vice versa—within a nine-year window have increased from 15 or 16 percent of income, up and down, to 17 or 18 percent.
Chapter 4 - Shaky Ground? Trends in the Risk of Large Earnings and Income Drops Among Demographic Groups
I examine trends in the risk of large earnings and income drops for several different demographic groups in order to determine the extent to which trends and levels of instability are accounted for by those variables. The results indicate that instability is highest among less-educated men and women, black men, new mothers, and women whose relationship ends. Over time, instability rose among men within each educational category, among less educated women, among men experiencing a breakup, and among families headed by an adult under age 40 or a college graduate. Differences in trends across demographic groups explain very little of the overall trends in instability observed in Chapters Two and Three. Rather, higher educational attainment and, among women, declining and delayed fertility and relationship formation shifted Americans into lower-instability demographics over time.
Chapter 5 - Conclusion: Economic Risk, Policy, and Politics
Trends in other indicators of risk (joblessness, health care coverage and affordability, retirement security, and indebtedness) confirm that economic risk has not grown substantially in recent decades. Economic insecurity has grown since the early 1990s recession, but evidence suggests that this may be due to disproportionate coverage of downsizing in the media. Levels of economic insecurity are comparable to levels of anxiety about other risks such as dying in a plane crash or becoming a victim of crime, and people generally say their own economic situation is much better than their assessment of others' situations. This chapter also illustrates flaws in Jacob Hacker's and Paul Pierson's Off Center, showing that the range of policies Americans will support is wide enough to incorporate (most of) the domestic agenda of George W. Bush and the GOP Congress of the 2000s. It is also wide enough to incorporate left-leaning policies, though Hacker and Pierson (and others) overstate the strength of support for such policies. Finally, I propose a system of "citizen benefits" designed to address economic insecurity while attracting widespread support and not blowing up federal deficits any further.
Appendix 1 - Previous Research on Earnings Instability and Volatility Trends
Appendix 2 - Previous Research on Income Instability and Volatility Trends
Files for Replication
The zip archive contains a number of STATA do files. Start by reading "master.do", which explains how to obtain my PSID extracts and other data, indicates which do files apply to which chapters, and explains how to set up the necessary files to obtain results.
