Stagnant Wages and the Financial Bubble

I’m happy to see that both Alyssa Katz and Kevin Drum also think that there’s a ‘there-there’ to investigating the Rajan hypothesis I blogged about earlier, and each kick out a version of the hypothesis that you should go check out.

But there’s an interesting dissent to the hypothesis. Scott Winship at Progressive Fix lays out data that goes against this argument, concluding “Again, the point here is that progressives should care about the facts.” Oh dear. This is with a hat-tip to Will Wilkinson, who says that the left should be reminded of “of the general importance of a reality-based stance.”

Now I’ve always been more of a buffalo stance kind of guy, but I know enough about the internet to know one has to defend his or her reality-based credentials. So let’s check out this argument.

I: Stagnant Wages to Credit Bubble

You should go over and read Winship’s argument, but here’s the key point which also includes Rajan’s argument:

And yesterday [Kevin Drum] endorsed the views of economist Raghuram Rajan, who blames the financial crisis on “the purchasing power of many middle-class households lagging behind the cost of living.”…in terms of the level of income the typical American has and in terms of recent trends, a careful look at the data implies that the middle class is doing pretty well.

First, consider what many progressives consider “the good old days”—the height of the pre-1970s economic boom. In 1973, the median inflation-adjusted income was higher than it had ever been and higher than it would be again until 1978—$45,533 (in 2008 dollars). Call this the gold standard before, in the conventional progressive telling, things started going south.

How much did things go south? Well, in 2008 the median was $50,303. That’s right—about $5,000 higher (after adjusting for changes in the cost of living).

So the argument from my side is that at some point household income went stagnant and taking on debt took over for it. What Winship points out is that wages are actually higher today in 2008 than back in a supposed golden-age of 1978. Wages are $5,000 higher in fact (inflation adjusted). Now a quick read of that would make you think wages had always been increasing throughout this time period. However checking the data that he provides, wages didn’t peak in 2008; actually they peaked in 1999 at $52,587, or over $2,000 higher than it was 9 years later in 2008. Let’s plot that data:

Median household income is increasing from the early 1980s until 1999 when it went stagnant. Mean household income also peaked in 1999 in this data source. So, anything interesting happening in changes to consumer borrowing around 1999? Check this out:

Huh. This is Michael Mandel’s Economic Statistic of the Decade, taken from here (it’s the derivative of total borrowing). That’s something, borrowing goes crazy right around the time median household income peaked. I’m going to take both charts and normalize them to their value in 1980 (click through to get a larger picture of any of the charts):

I’m on a time crunch, so I’m sorry that’s a terribly messy graph, but it gets the point across – income flattens out (right y-axis) right around the time borrowing takes off. There’s clearly a “there-there” to the Rajan hypothesis in the graph. Now what is it telling us? I don’t know. Maybe the causation is backwards and people are working less cause they feel so rich off their houses. Maybe the growing inequality during this time period plays a role, or maybe it doesn’t. Same with the financialization of the economy. Maybe the middle-class suffers from the disease of ‘affluenza’ and buy too much stuff they don’t need, though Mandel found that couldn’t be the entire story. Maybe this is all just an unhappy coincidence. But this certainly needs more investigation, and shouldn’t be dismissed out of hand.

Winship doesn’t technically have an answer to this, other than (I think) the middle-class should feel richer because these income numbers don’t have the proper deflator on them. I’m not an expert on PCE versus PCI (who is? drop some comments), but is this one of those “your wages won’t increase but Wal-mart has very cheap goods so feel rich” arguments? If you are like me and view money as something that buys both consumption goods and autonomy in a liberal capitalist democracy, then the consumption good deflator only covers half the reason of why a worker values getting a dollar after a day of work.

II: Wage Growth

As for the increase in household median income from the late 1970s to 1999, I’m under the impression from many sources that the majority of it is simply households working more hours, particularly by getting a second worker into the workforce. That’s right, right? Here’s Steve Ross who is quoted as an excellent source in the critique for middle-class numbers:

It is true that much (but not all)* of household income gain can be attributed to wives working more, but neopopulists see the increased female workload in an entirely negative context and as a burden on women and families. We suspect many working women want to work….* If the working hours of wives are held constant at 1979 levels, median incomes at the 50th percentile for prime-age couple households would still have risen by 9 percent in real dollars from 1979 to 2004. For households at the 70th percentile, the increase would be 22 percent.

I’m happy many more women are in the workforce. But if the real driver of household wage growth is simply that households are working more hours, then that isn’t exactly wage growth. “I want to make more money so I doubled my shifts” isn’t really the same statement as “they are paying me more money.” Winship points out male wages have increased around 8% over a 35 year period, for an annual growth rate of 0.23%. Something eventually had to give: maxed out at capacity, both in the number of hours in a day and the line of credit approved – if this was a firm, would you want to invest?

III: Middle Class Woes

As a last aside, his analysis saying that the middle-class is doing fine surprisingly misses the biggest story that’s happening to them: foreclosures, underwater mortgages, and the housing crash. From The consumer isn’t overleveraged — the middle class is:

What’s more, on the asset side, BofA Merrill says the middle-class has suffered more than the wealthy from the housing crash because middle-class families tended to rely more on their homes to build savings through rising equity. Also, the wealthy naturally had a much larger and more diverse portfolio of assets — stocks, bonds, etc. — which have mostly bounced back significantly this year.

We talk about that more here. The rate of foreclosures won’t peak until the end of 2010, and will remain high afterwards. Meanwhile you have a huge amount of the middle-class underwater on their mortgages, and perhaps banks that can’t write down the LTV to market rates without going insolvent, leaving everyone in a standoff where the only options left for people is to keep throwing money down a failed investment (“How I Learned To Stop Worrying And Love The Sunk Cost Fallacy”) and keep bloated LTVs on the banks’ balance sheet or walk away from their mortgages. These are terrible options that call for at least some progressive thought thrown behind them.

This entry was posted in Uncategorized. Bookmark the permalink.

22 Responses to Stagnant Wages and the Financial Bubble

  1. Rebecca Z. says:

    Didn’t “median household wage” stay up because wives went to work?

    Wouldn’t this mean that the purchasing power of an individual wage earner dropped?

    • Rebecca Z. says:

      excuse me, I wrote too soon.

      thank you for pointing out the apples-to-oranges logic of household wage as a measure for this time frame.

  2. pushmedia1 says:

    For section one, I’m not sure what comparing a level to an median tells us. I think you want a version of the last chart in the post but with trends over time, i.e. does the debt load of the middle class increase after wages stagnate?

    For section two, if you correct for the decreasing size of the household, you’ll get some (very little) growth in per household member wages. Also, your choice of deflater will change your result. I like the PCE in this context because you’re interested in how much stuff a middle class wage can buy.

    For section three, I don’t get this analysis. For every middle class home owner losing his shirt, there’s a middle class home buyer that’s getting a lower price for a home.

  3. pushmedia1 says:

    I should have said: if you correct for the decreasing size of the household, you’ll get some (very little) growth in per household member wages OVER THE LAST DECADE.

  4. RW says:

    I am rather surprised Elizabeth Warren is not being cited here since she is the recognized authority WRT hollowing out of the middle class in the United States. Here is a video of a presentation she gave a few years ago that outlines her research and main conclusions –

    At the risk of bowdlerizing her position, the main point seems to be that (a) fixed, non-discretionary family expenses have gone up much faster than family income and (b) there is virtually no economic cushion in the two-wage earner household so additional increases in those costs must increase the families need for leverage.

    Note: Dr. Warren adjusted all her data for inflation but wisely chose to work with disaggregated Commerce Dept and Labor Dept data and so avoided comparisons to mean cost of living metrics that would mask the distinction between discretionary and fixed family expenses, tend to underestimate large fixed costs such as housing, and could not facilitate investigation of family expense allocation (where the money is actually going).

  5. Dismal says:

    What’s your opinion on the use of household as opposed to per capita income?

  6. retr2327 says:

    “If the working hours of wives are held constant at 1979 levels, median incomes at the 50th percentile for prime-age couple households would still have risen by 9 percent in real dollars from 1979 to 2004. For households at the 70th percentile, the increase would be 22 percent.”
    And I take it from your that the overall average increase was 8%. Now, I’m not much on math, but that data set would seem to imply that those below the 50th percentile got slaughtered.

    Trickle-down indeed.

  7. Kaleberg says:

    i’ve been curious about this myself. I did an analysis of how many hours of work at the median wage it takes to pay the mortgage on a median priced house at the prevailing mortgage rate. It was about 600 hours through the 60s. It soared in the late 70s, but settled about 800 hours in the 80s and stayed near there into the recent bubble. That change overlapped with the rising numbers of women in the work force in the 70s and 80s. [see

    I also took a look at the shrinking fraction of the goods and services our nation produces that one can command by working a full time job. Interestingly, this has been falling since 1960, so the pay from a full time job lets you command only half as much of your share of the GDP as it did back then. I have no idea of what this all means. Wages are still only half of all personal income despite this change. Is personal income that much a smaller share of all income? [for more confusion or

    I would love to see some economist make some sense out of all of this, but so few of them seem to notice that for most people, the economy is the money they get by working and they spend on stuff.

  8. vimothy says:


    How do you see the Clinton-era budget surplus feeding into this trend? (Since G – T < 0, it's effects ceteris paribus must have been recessionary in exactly the same way that increased private sector savings rates during the current crisis have been recessionary, and could plausibly have contributed to the build up of private sector debt. (Indeed, scholars at the Levy Institute made this very prediction back in the late '90s)).

  9. Pingback: More on Wages and the Middle Class: A Response to Rortybomb | Progressive Fix

  10. Pingback: High income inequality hurts political engagement « The United States of Jamerica

  11. So Winship points out that median household income has gone up a little since 1978. But as you note wives are working far more hours and husbands are working a lot more hours too.

    One thing that this means is that there’s much less in-home labor that should be counted in GDP but isn’t. If it were counted then the median household income in 1978 would be a lot higher.

    But what are the effects on expenses? They’re a lot higher. When I grew up few families paid for daycare at all, because there was a stay at home wife. Eating out costs were way lower with a homemaker to cook. Driving costs were way lower, maid service and landscaping services were rarely used, and so on.

    In addition, government transfers to the poor and middle class were much greater. What a family had to pay for college was far less due to far greater government support paid for by far more progressive taxes. The top marginal rate was cut in half in this period.

    Income risk and so many other risks also increased greatly.

    And one last point, although if not for time I could go on. Cornell economist Robert H. Frank makes a very important point about income inequality increasing causing positional/context/prestige expenditure cascades. People did not want to lose their position with the wealthiest pulling away and this pushed them to increase their spending too, turning to debt to not fall far behind.

  12. This site ( lists the average daycare cost at $8,150 per child per year, according to data from the National Association of Child Care Resource & Referral Agencies (NACCRRA).

    For two children, that’s $16,300. That alone dwarfs Winship’s $5,000 increase in median household income from 1978 to 2008. Women (and men) having to work so much more to get that meager $5,000 increase over 30 years meant they were working in the home far less and having to spend the $16,300 on daycare, and the thousands, or tens of thousands, more on eating out, much more driving, hiring people to help with landscaping, cleaning, repair, finishing the basement, etc. All of that home work which was done by a stay at home mom and a dad who worked far less was, in fact, work, but it’s a well known flaw in economics that it’s not included in GDP. Had it been, I’m sure you would see a clear fall in the median income from 1978 to 2008.

    But look, a bigger point is that even if the median household income actually did increase by a few crumbs, the 1/5 of 1% per year that Winship crows about, there’s no reason why we have to be happy with this when almost all of the growth is going to the wealthy and super wealthy. It’s just not necessary to settle for such low growth for the vast majority. There’s an easy alternative, more progressive taxation and far more high return investment in the vast majority, in things like education, college aid, free universal pre-school, universal health insurance, smart infrastructure, alternative energy etc. (for more on this please see: These investments in the people and the country not only result in far more even growth, they result in a much bigger and faster growing total pie

    There’s no reason we have to settle for this. We’ve seen from history before the Republican era what relatively progressive policies can do. As Paul Krugman writes in his 2007 book, “The Conscience of a Liberal”:

    During the postwar boom the real income of a family roughly doubled, from about $22,000 in today’s prices to $44,000. That’s a growth rate of 2.7% per year. And incomes all through the distribution grew at about the same rate, preserving the relatively equal distribution created by the Great Compression.

    End Quote

  13. Oh, the quote from Nobel Prize winning economist Paul Krugman is on pages 54-55.

  14. Here’s some more on Winship:

    He writes:

    “what has likely happened is that the very top—the top one-half of one percent—has pulled away from everyone else, though the increase from 1980 to 2009 has probably been fairly modest.”

    How can he write, “though the increase from 1980 to 2009 has probably been fairly modest.”?

    If you look at the gold standard data of Berkley economist Emmanuel Saez (the “TABLES…” link in the first paragraph of his home page at:, you see that for the top .01% their income increased from 128 times the average in 1980 to 604 times in 2007 (the most recent data). For the top 1% the increase was from 10 times to 23.5 times.

    And this is before taxes and transfers, and since 1980 tax rates have plummeted on the wealthy thanks to the Republicans, making the gains by the very top even more extreme. The top marginal tax rate in 1980 was 70%. Today it’s half that, at 35% (see:

    So, it’s ridiculously untrue that their “increase from 1980 to 2009 has probably been fairly modest.”

  15. He also talks about how spending on employee benefits has increased.

    With regard to health care, or health insurance benefits. The dollar value of health insurance benefits adjusted for inflation may have increased, but that’s because the cost of health care increased so much faster than inflation.

    In fact, the percentage of people without health insurance is much higher than it was a generation ago, and the co-pays and deductibles that must be paid are far higher. So adding benefits as a factor just makes today’s families poorer, not wealthier. It makes the stagnation worse, not better, or it turns it into decline.

    Of course, the quality is better with a generation of technological advance, but the key point is that there’s no way we should be forced to settle for a country where almost all of the income advance (or more than all of it) goes to the wealthy. It’s just not necessary. Before the Republican era, the pie itself grew much faster and the gains were pretty evenly distributed. So there’s no reason why the vast majority of Americans have to settle for crumbs or losses with all of the gains going to the wealthy.

  16. And my last experience with Winship made me suspect:

    I really don’t like what I’m seeing from this “Progressive” Fix blog. They’re really starting to look like conservatives, or at least far from progressives, posing as progressives.

    Look at this post:

    They show a chart on government spending projections spiraling out of control by 2075 that’s almost completely driven by health care prices rising much faster than inflation and being assumed to continue to do so forever. They conclude:

    “But let there be no doubt — the long-term prospects for significantly expanded progressive government are dim, and in fact, a retrenchment in coming decades is inevitable. President Clinton was wrong — the Era of Big Government is not over. But it will be soon. As progressives we must lead the process of winding it down in a responsible and fair way.”

    What is this “we”? You certainly are not progressives here. If health care prices keep increasing at the same rate forever, not only will government eventually have to be slashed, so will everything else government and private. What conservatives want, what everyone wants, will have to be slashed, slashed to zero because health care will be 100% of GDP! The chart doesn’t say that progressive goals are impossible, it just says any goals are impossible if health care costs keep growing at the same rate they have been indefinitely. But, of course, they won’t. Eventually we will come to our senses, or they will get so high there will be a voter revolt and stern action. Hopefully, strong action to control health care cost growth will happen very soon with the Democrat’s plan passing.

    By the way, the Earth is doomed anyway. I gained 5 pounds over the holidays, so that means in long run projections I will weigh so much I will destabilize the Earths orbit.

    • Nancy Irving says:

      “…because health care will be 100% of GDP!”

      And no guarantee it will stop there–we can always continue to pay *more* than total GDP on health care, as long as the Chinese continue to buy our bonds! 🙂

  17. With regard to my last comment,

    The Baseline Scenario has a graph today that illustrates it perfectly at:

    Of course with the news of the last two days everything seems dwarfed, or dependent on, abolishing the filibuster, and then with that, establishing very strong public campaign finance. For more on this see:

  18. Pingback: 8 Reasons the US has Become a Nation of Slaves | Wall St. Cheat Sheet

  19. Pingback: Real Wages & Wealth Disparity « Marmalade

  20. Pingback: Collective Conscious » Why Demand Money?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s