Forecasting HH Income

December 22nd, 2011 at 1:35 pm

Someone asked me what I how I thought middle-class incomes held up in 2011.  We won’t know until Census releases median income numbers for 2011 in the early fall of 2012, but we can make an educated guess.

That is, based on the time-series of household median income going back to the mid-1960s, along with a bunch of correlates for which we have 2011 data, one can make a somewhat reliable forecast (with a forecast error of around 1% in terms of nominal yearly growth, which ain’t nothing, but there it is—more nerdy stuff on the forecast below).

So, here’s the graph of actual and forecasted nominal—before inflation—median household income since the mid-60s.  As you can see, the forecast hugs the line pretty tightly, though something a little funky seems to be happening at the end of the series, and that’s the part we care most about right now.


Source: Census Bureau for nominal median household income data (HHINC); my calculations for forecast (HHINCF) as described in text.

What happened there was unique and actually important in terms of understanding the depth of the squeeze on middle-class families.  For the first time in the HH income series, median household income fell— in nominal terms, i.e., before inflation— not once, but two years in a row.  Now, in 2009, inflation actually fell too, so there hat was pure deflation upon the land.  But in 2010 inflation went up and same with this year (2011), probably by around 3%.

As you can see, I forecast median household income to be flat in 2011, before inflation.  Even if I’m low by a few percent, it’s hard to imagine there was much of a pop on the income side for middle- and lower-income families in 2011.  It’s just been a slog of a year.

There are some signs that things are picking up a bit.  Growth is expected to be pretty strong in the current quarter, some recent housing numbers suggest that market is finally carving out a bottom, and jobless claims have fallen consistently in recent weeks.  But until the job market starts posting consistent gains of a greater magnitude then we’ve seen thus far, middle class families will remain squeezed.

Policy implications of all of this: with a little bit of mo here, this would be a particularly lousy time to let the UI and payroll extensions expire.

[Data note: to generate the forecast, I regressed nominal HH income on inflation, weekly earnings, aggregate hours worked, and the lagged dependent variable, with all variables entered as log changes.  Since I used average annual values, I had to guesstimate Dec values for the three independent variables.  The forecast is “static,” meaning I used actually values for the lagged DV, not forecasted values.  That gives a more accurate prediction but restricts you to one-step-ahead, i.e., nothing beyond 2011.  Note that the 2011 forecast mostly excludes the impact of the payroll tax credit (it may have found it’s way into the weekly earnings regressor, but not sure), so it is probably biased down a bit.]


Print Friendly, PDF & Email

2 comments in reply to "Forecasting HH Income"

  1. Dan Furlano says:

    John Carney, senior editor CNBC, on MMT.

  2. Jean says:

    So I have a question, that isn’t really related to this post, but here it is anyway …

    If the Government had bailed out the Mortgage Crisis through home-owners instead of through AIG, would that have made as big of a difference as I’m thinking it would?