Huffpost Politics
THE BLOG

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors

Jared Bernstein Headshot

Poverty, Income, and Health Insurance: What to Expect and Why It Really Matters

Posted: Updated:

Every year around this time, the Census Bureau releases one of the most important government reports: the annual status of poverty, household income, and health insurance coverage. The release is scheduled for the morning of August 26th at 10; you can usually watch the press conference over the web if that kind of thing turns you on as much as it does me.

With your approval, I'd like to take you on a pre-release tour of these data, discussing what to expect -- slightly lower poverty and higher median income; less health coverage -- and, more importantly, the context within which to interpret these results.

While 2007 probably seems just so last year, the release is uniquely important...even historical. That's because 2007 was almost certainly the last year of the 2000s recovery, and the Census release enables us, for the first time, to evaluate how the living standards of middle- and low-income families fared over this recovery. It's not likely to be pretty.

What's that? You already know how they/you did? Good point. But this makes it official, and the release, which generally gets front page status for a day or two, gives us a chance to reflect on an economic expansion that left most people behind. It also happens to be a presidential election year, with the economy front and center in precisely the manner invoked by these data. As I've said in this space before, the candidate who understands this disconnect between growth and living standards, and whose ideas are best crafted to reconnect them...that's the one who should win.

The first thing you should know about these data is that they refer to calendar year 2007, as the Census Bureau treats insurance coverage and poverty as annual concepts for this report: if your family income was below the official threshold last year, or you went without any kind of health coverage, public or private, for the whole year, they count you as poor or uninsured.

There are known glitches: the poverty concept is way outdated, and is generally agreed to understate the degree of actual material deprivation. In 2006, 12.3% of the population -- 36.5 million people -- were officially poor. But when a couple of statistical scholars put together a more comprehensive, alternative measure, they found a 2006 poverty rate of 17.7%, adding 16 million more persons to the poverty rolls compared to the official measure.

Also, the income measure is pretax, and since taxes have been cut a lot in recent years, a post-tax measure would show more income growth, though this is less important for the middle-class relative to the rich, whose taxes have been cut the most.

So, with all that throat-clearing out of the way, here's what to look for. Wait...one more caveat. When it comes to forecasting economic reports, econometricians are no better than weather-persons. So, if I get these right, tell everyone. If I'm way off, we never spoke.

• Poverty probably fell a bit, I'd say to 12.1% (from 12.3% in 2006).
• Median household income, adjusted for inflation, probably rose about 1%.
• There are likely more uninsured people, especially due to lost employer coverage.

If the economy's in the tank, why do I think things improved? Because for most of 2007, the economy wasn't in the shape it's in now, and usually -- not always -- in the last year of an expansion, poverty goes down. The key determinants tend to be jobs, wages, and inflation, and all were actually in decent territory until the last quarter of last year. However, things have of course deteriorated since -- we'll get to the 2008 story in a moment.

But here's the first contextual kicker: assuming I'm right about the direction of these results, before anyone uncorks the champagne over the first two of them, consider these facts. Compared to the peak of the last business cycle in the year 2000:
--a larger share of the population will be poor;
--real median household income will be lower;
--there will be millions more people without health coverage.

(I'm sure about #'s 1 and 3. I could be wrong re #2, but if real household income surpasses the 2000 level, it won't do so by much.)

Remember folks, over these years, productivity was up 19%, growing faster over the 2000s than in the 1990s (on a yearly basis, productivity was up 2.5%, 2000-07, vs. 2% per year in the 1990s; that little 0.5% difference actually means a lot over a few years). The mantra among economists is that as grows productivity, so shall living standards. But in the 2000s, most of the American workforce worked harder and smarter, yet ended up earning less. Looks like we need a new mantra.

Here's the second kicker: last year was as good as is gets for awhile. The historical precedent is extremely clear on this score. In recessions, poverty rises and median income falls. And it's not just recessions. The last few recoveries started out as jobless, meaning the economy as a whole was expanding, but employment was still contracting. Poverty rose and incomes fell for the first few years of the last two recoveries.

The reason for these weak-start recoveries is that labor market conditions are a primary determinant of working/middle-class living standards. These families depend on their paychecks, not their stock portfolios, and with jobs and real wages in decline all year so far, the clear expectation is that middle-class incomes are down in 2008, and poverty is up.

Plugging current conditions into the models, I get poverty back up to 12.4% in 2008, and real median household income down around 1%, a loss of about $600. (For those interested in the sausage-making process, this assumes inflation subsides in the second half of this year; if it stays elevated, that $600 loss could become a $1,000 loss).

Going back to the 1940s, we've never completed an economic expansion where the middle-class family income failed to regain its prior peak (note the subtle switch from 'household' to 'family' income -- the Census family series goes back to the mid-1940s; the household series only goes back to the mid-60s; the main difference is that families exclude one-person units). I can think of no more damning indictment of the current economy -- no better example of how it is broken.

That's why next Tuesday's release is such an important one. True, it's last year's news, but if I'm anywhere close regarding these forecasts, the results will shine a red-hot spotlight on the biggest challenge we face: the disconnect between growth and living standards.

Look for the candidates to release statements based on the findings. Both will distance themselves from the Bush policies that have surely played a role in the disconnect. The problem for McCain is that his economic agenda seems to have been formed by looking approvingly at results like these and deciding the best path forward was to double-down on Bushonomics. He extends the Bush cuts for the wealthy, and throws in a big corporate tax cut for a cherry on top. His health care plan will be much less effective in covering the uninsured.

Obama (for whom I'm an informal advisor) starts with tax relief targeted at the middle and lower income families. After-tax income grows most for the poor under Obama, and most for the rich under McCain (see figure 1 here). Obama offsets the disconnect; McCain exacerbates it.

But tax policy can only offset the disconnect; it can't repair it. For that you need robust job creation, restraints on capital excesses, deep infrastructure investment, single-payer health care, and long-term energy policy with job-creating investments in green tech and alternative energy sources.

So, I hope to join all of you on the Census Bureau website next Tuesday, August 26th. I recognize that my forecasting prowess is on the line, but somehow, I think the stakes are a little higher than that. The release and the analysis surrounding it should provide a close and critical look about where we've been and we're we need to go.