The Source of the Statistics and How They Were Analyzed
While most economic statistics involve taking snapshots at different points in time and comparing them, researchers have developed “through-time” measures that shed more light on how events actually unfold.
One such measure, used to gauge risk in the stock market, is called a “beta.”
The beta of a stock is how much its price rises and falls relative to changes in the overall market as measured by an average like the Standard & Poor’s 500 Index. The bigger the price swings, the greater a stock’s risk - that is, the more likely a shareholder will lose money if he or she has to sell at a particular moment.
In effect, the Los Angeles Times has adapted the concept of beta to family income, measuring the changing risks that working Americans face.
The Times used the Panel Study of Income Dynamics for its analysis of family income volatility.
The panel study has followed a nationally representative sample of about 5,000 families and their offshoots for nearly 40 years and is the most comprehensive publicly available income and earnings database in the world. It is run by the University of Michigan and principally underwritten by the National Science Foundation. The families’ identities are kept confidential.
The Times employed techniques for gauging income volatility that were developed by economists Robert A. Moffitt of Johns Hopkins University and Peter Gottschalk of Boston College. The Times also consulted with Yale University political scientist Jacob S. Hacker, who has conducted his own analysis of income volatility among panel-study households and published a paper linking it to economic risk.
The Times employed two Johns Hopkins graduate students, Xiaoguo Hu and Anubha Dhasmana, to help generate the data. Moffitt guided them and advised the newspaper.
The Times’ analysis looked at five-year increments from 1970 to 2000 and examined the annual fluctuations in each family’s income.
For example, for a family whose income rose from $40,000 to $45,000 over a five-year span, the paper examined the journey from the lower number to the higher. Did the change occur in steady $1,000 annual increases? Or did the family’s income take a big jump in one year and plunge in another?
The Times’ basic finding is that the fluctuations in annual income that individual families have experienced have grown larger over the last three decades.
Based on the panel-study sample, The Times estimated the annual income swings, up or down, for 68% of all U.S. families - those who did not have the most extreme fluctuations. As a result, the newspaper’s conclusions don’t rest on cases outside the mainstream: the movie star whose career dries up overnight, say, or the hourly worker who wins the lottery.
To zero in on working families, The Times focused on men and women 25 to 64 years old whose households had some income.
The analysis looked at pretax income of all family members from all sources, including workplace earnings; investments; public transfers such as jobless benefits, food stamps and cash welfare; and private transfers such as inheritances.
All amounts were adjusted for inflation, expressed in 2003 dollars.
For a family to be included in its sample for any five-year period, the newspaper required that there be records on that family for at least three of those years. It maintained this standard even during the final years of the 1990s, when the panel study changed from annual collection of data to an every-other-year schedule.
To avoid double-counting income in cases where families broke up, The Times added such payments as alimony and child support to households receiving them and subtracted the same amounts from those paying.
The newspaper ran a series of tests to determine whether its results were the product of data problems or reflected real changes in the economy, and to address concerns about some aspects of the panel study that have been raised by specialists.
The Times was forced to eliminate a sub-sample of Latino families that panel-study researchers sought to add in the early 1990s. Database officials and outside researchers have concluded that the sub-sample was poorly designed and that its inclusion would distort results.
The paper ran its numbers with and without a separate immigrant sub-sample, as well as with and without adjustments for changes in the size of families.
It ran them with and without transfer payments, which some researchers believe were poorly recorded for several years during the early 1990s. It also ran them with and without data for 1993, the year about which these researchers have the greatest concern.
In order to check whether its results were the product of a few outliers whose incomes either rose wildly or fell precipitously rather than a broad trend among most families, it ran the numbers trimming the top and bottom 3% of families.
In every case where the results with these changes were similar to those without, the paper added back the removed families in order to make the fewest possible adjustments to the database and to maintain the largest possible sample.
