Employment and wages

Russ,

Your comments are correct on the impact this trend has had on average wage rates.


However, the more important point relates to the underlying cause: the floating dollar.


Once the dollar lost all ties to anything real, it led to a decline in the living standard, forcing spouses (usually women) into the workforce just to maintain the standard of living, and destroying families in the process. The floating dollar creates mixed signals that distorts production, labor, and family.

The reason for the huge increase of women in the labor force; The Pill.

Regardless, I am surprised that anyone would suggest "The biggest reason is the mismeasurement of inflation which leads to an underestimate of earnings." The statement puts you squarely on the side with about... no one. Underestimating the increase in buying power of wages over the last 40 years? Mayhaps you have access to the secret group who tracks real inflation? Every shadow group that I know records inflation at rates much higher than the official figures. Now you claim the offical figures are too high?

Rob Dawg,

Many studies of the CPI (Hausman, Eisner and Boskin are the most prominent names) estimate that the CPI has been overstated by about 1 percentage point a year over the last few decades due to a failure to control adequately for quality improvements. Over thirty years, that's means that if measured standard of living are stagnant, the actual increase is about a 30% improvement. That's an enormous difference.

Actually, Russ, if you look at labor force participation rates of married women, you see a very steady growth pattern from, say, 1967 to 1992. There is no appreciable blip up during the 1970s. See the scan of two charts from Joyce Jacobsen's The Economics of Gender that I've put here.

Your case is stronger if you decompose "married women" into "married women with their youngest child under 6." That group did see a blip upward in the longer-term trend line starting about 1972 or so. And it may be that those women were especially likely to work part-time and/or in jobs with flexible schedules where monetary wages were lower.

And the case for more divorced women in the labor force in the 1970s is a good one. Those numbers jumped indeed.

There is also substantial evidence for the claim that many of the women who found themselves working in the 1970s did so unexpectedly, hence many had not acquired the human capital that would command higher wages. In 1968, according to the National Longitudinal Survey of Youth, 28% of white women age 14-24 planned to work at age 35. In 1980, more than 70% of those same women were in the labor force. Those who were in the labor force but hadn't planned to be likely fit your story of lower than average wages for new labor force entrants.

The Boskin Commission in 1996 found three major problems with the CPI.

To address these critisms the BLS changed its methodolotgy in three ways.

One, they shifted to the rental equivalence to measure the cost of shelter. Although since that change other measures of housing prices have risen significantly faster then the rental equivalence reported. Since this is the single most important factor in the index this has caused the index to significantly understate inflation in recent years.

Two, they changed to use a geometric mean methodology to capture consumer substitution bias so when consumers changed their purchases because the price of one item like beef rose they replaced it with chicken.

There is still considerable debate about this change because if rising beef prices cause one to buy chicken it is a drop in


living standards so using this methodology


may bias the index downward. Under this methodology we could all be reduced to eating Solvent Green and it would not show up in the data as a lower living standard.

Third, the BLS shifted to hedonistic measurement methods to capture quality changes in the CPI.

So in recent years the reported CPI actually incorporates essentially all of the problems you claim cause the CPI to overstate inflation.

Yes, the BLS did not go back and change the history of the official CPI series. But they did start publishing a new series, the CPI-RS, or Research Series. This series did go back and recalculate a history of the CPI

incorporating the new methodologies.

Moreover, Census, and other official data collectors have shifted to using the CPI-RS

when calculating the history of such things as real wages and income.

So, for example all official measures of real average hourly earnings is based on the CPI-RS, not the official CPI.

So when are you going to quit ignoring these developments that have essentially corrected the data for the criticisms in your analysis.

"Real average hourly earnings as measured by the BLS peaked in the United States in the late 1970s leading some to conclude that the average worker has made no progress."

I've been banging on about this for a couple of years now, as has Alan Reynolds (obviously, he banging on about it has more influence than I do).

The commonly quoted BLS number is all hourly paid workers. Over and above the rise of women in the workforce there is also the change in composition coming from more part time worker. Both women and part timers get less per hour worked that full time men do.


What we'd actually want to look at is hourly pay for male full timers....don't know if the BLS calculates that on a regular basis.

Russ,

Your point about hedonics and its slippery character is well taken. At the same we've had 2 major and a great many minor adjustements to the BLS measure of inflation over the same period. Nearly all have had the effect of overwhelming any hedonic benefits with the reported downward changes to CPI.

No doubt the average two earner family is living a bit better than the average single earner family of 1970. Can I pay California In-State fees* as adjusted by CPI and get my kids into UCLA? What is $320 adjusted from 1970 to 2007? BLS CPI Ans: $1714.78. Truth: $ 7,713.23

* Fees: The California Constitution prohibits charging "tuition" so they charge "educational fees" instead.

Regarding the idea that workforce participation of women was increased by the rising divorce rate, is it possible that you have the cause and effect reversed? Perhaps an increased participation of women in the workforce caused an increase in the divorce rate since more women were capable of financial independence. Or perhaps the cause and effect went in both directions.

Wouldn't the entry of women into the workforce raise the wage rate if all work were counted? Housewives definitely are workers, yet generally have $0 wages. As these women move to the paid workforce, low wage service occupations increase. Housekeepers, nursery school attendants, fast food workers, laundry and dry cleaners workers, and seamstresses - all working at near-minimum wage - now perform tasks formerly done by $0 wage housewives.

If we believe in comparative advantage among the female population, we should all be better off. Yet official statistics may show an apparent wage decline.

Spencer,

You're right, the BLS made some changes in response to the Boskin commission. But the most important problem, the issue of quality is not fixed. Yes, my understanding is that there is SOME adjustment for quality. But my impression based on the work by Mark Bils (cited here at the Cafe many times, anyone interested, please search) is that it is grossly inadequate.

But I will try and find time to get a better idea of what exactly the BLS has done on the quality front.

For John Dewey: yes, if all work were counted. But of course housewives/husbands do not get counted as a $0 wage, but as not in the workforce.

For Dave: there is some evidence that women begin to increase the hours of labor they supply in the two to three years before a divorce, presumably in anticipation of that divorce and the need to become more independent.

More generally, I think it's a reasonable argument to claim that the demand for unilateral (no-fault) divorce was an effect of women's improved economic circumstances. More women could survive outside of a bad marriage and thus they wanted ways to make it easier to get out. To the degree that no-fault led to higher divorce rates, then it does seem plausible that women's increased LFP might help explain higher divorce rates. The causality here runs both ways.

I highly recommend Claudia Goldin's 2006 Richard T. Ely lecture in the May AER as a one-stop source for many of these issues.

steven horwitz: "But of course housewives/husbands do not get counted as a $0 wage, but as not in the workforce."

If I wanted to do so, would it be possible to determine the wage rate for all workers - both the paid workforce as well as the unpaid workforce?

The wage impact of moving "low skilled" tasks such as food preparation, housecleaning, child-raising, etc, from the unpaid workforce (or non-workforce) to the paid workforce must be significant. Surely someone has analyzed this, haven't they?

Sure it's significant - that's a good hunk of the work women are doing, especially immigrant ones. As the opportunity cost of women staying at home to do those tasks rose, they entered the labor market and paid other women (and men) to do it for them. It's the virtuous cycle of growth.

OK. Thanks, Steven. It helps to hear from an economics professor that I'm thinking correctly. It's been 25 years since my last economics class.

Over the past decade the cpi home owners equivalent rent has averaged a 3.3% rate of gain. But over the same time the OFHEO House Price Index, of repeat sales showed a 11% average annual increase.

Give that home owners equivalent rent accounts for 22% of the CPI this implies this factor alone has caused the CPI to understate inflation by 1.8% percentage points each year.

By comparison Mark Bils outside estimate is that inadequate quality adjustments have caused the CPI to overstate inflation by 0.6

percentage points annually.

When you find another 1.2 percentage points over statement of the CPI let me know.

I love the use of hedonic adjustments to correct for improvements in the quality of products (goods and presumably services). But do they apply them in reverse? If you've flown recently, you know exactly what I mean.

I am impressed by how many intelligent people there are participating on this blog. It makes for interesting and educational reading for me, and I bet it keeps Don and Russ moving to stay sharp enough to be considered as "out in front."

As to the numbers of divorced women in the workforce increasing in the 1970s there is another explanation other than the pill (definitely a factor) and women discovering that they were financially capable without a man. My explanation garnered from street observations and experience is that the breakdown in the standards of morality and responsibility that came to fruitition with the Baby Boomers is a huge factor. And, that generation entering the workforce occured slightly before that time.

I lived through those years and what I saw going on around me was large numbers of my young co-workers were married men who had married their HS sweetheart and most had at least one child right away. Their wife may or may not have been a stay at home mom, but most of them never really shed all that weight they gained in pregnancy. So now my co-worker had an unattractive wife at home that devoted more time to the kid than to him. Not being the center of attention was unacceptable to a Baby Boomer.

At work he found himself rubbing elbows with women who may not be as attractive as his wife was when he married her, but they were now, and they were devoting attention to him. It didn't matter to him if the lady was single, married, or divorced, he had been convinced that if it felt good, do it. So, romance/lust created a freshly divorced mother with one kid. In short order to support the kid, she shaped up as much as possible (or not) and joined the workforce, where in many cases she became the other woman within two to three years, and another freshly divorced woman was looking to join the workforce. Mix those divorcees in with a nice number of sweet young things that were single but were looking for a good hard working man, and your numbers of women in the work force had to expand, and did.

Obviously not a scientific analysis and very much anecdotal, but it is what I saw, and still see to this day.

Married couples ceased remaining married, for the children and image in the community, and began divorcing in large numbers as the "me" generation entered the workforce.

The chart was interesting. Another thought on the rise in the mid-70's ... besides increasing divorce rate, birth control pills and possibly the cause of the increase in working married women ... the Arab Oil Embargo of 1973. After that October (I believe that's when it was), the price of virtually everything went up about 50 per cent ... almost overnight. I was in graduate school on a finite budget and remember it well. I'm surprised that no one mentioned it.

REB from GA,

I'm confused about the point you are making. Are you saying that women's participation in the workforce increased because higher oil prices reduced household incomes? and the women had to work to make up the shortfall?

Bruce Bartlett provides a sensible explanation for the increased participation of women in the 80's and 90's:

"It should come as no surprise that tax policies affecting the marriage penalty have had a significant impact on female labor supply....Analysis of the Tax Reform Act of 1986, which lowered the top marginal tax rate from 50 percent to 28 percent, shows that married women responded more strongly than men to the increased work incentive."

The Marriage Penalty

Nada Eissa, Faculty Research Fellow at the National Bureau of Economic Research , analyzed the labor supply responsiveness of married women to changes in the tax rate. She concluded:

"I find evidence that the labor supply of high-income, married women increased due to the Tax Reform Act of 1986. The increase in total labor supply of married women at the top of the income distribution (relative to married women at the 75th percentile of the income distribution) implies an elasticity with respect to the after- tax wage of approximately 0.8. At least half of this elasticity is due to labor force participation."

Taxation and Labor Supply of Married Women

I remembered another reason why women often earn low wages. For some married women, the purpose of a job is not to maximize income. In other words, they could be earning more but choose not to. Here's four real-world examples I've observed first-hand:

1. a 60-year-old well-off and well-educated widow worked at my bookstore for low wages in order to socialize with the charming male customers;

2. a middle-aged housewife infrequently sold real estate in order to justify and deduct her Mercedes;

3. some flight attendants, under union negotiated contracts, trade away nearly all their assigned flights but still enjoy free flight priveleges and paid vacations;

4. the wife of a highly successful surgeon opened a small European import shop that struggled to break even but allowed her to justify and deduct frequent trips to Paris.