Nate Silver runs some numbers on how various income brackets have fared under different Presidential administrations. He makes a very good point that is rarely appreciated:
This latter characteristic is something that I think a lot of liberals tend not to have a good appreciation of. There is sometimes a tendency among liberals to see the economy as a zero-sum game, but this is not really the case. When the economy is doing well, everyone tends to do well, unless the President is trying really, really hard (as Reagan did) to steer that growth only toward certain income classes. And when the economy is doing poorly, everyone tends to do poorly. The poor did awfully under George W. Bush, but the wealthy didn’t perform all that well either.
[emphasis mine]
The broad point is an excellent one, but I don’t think the attack on Reagan is justified by the data. In Silver’s data, he lumps together the performance under the Reagan administration and the Bush the Elder administration, and the combined numbers do indeed show household income growing for all income brackets, but growing much more for the upper brackets than for the lower brackets. By stark contrast, under the Clinton administration, income for all household income brackets grew roughly in unison.
But that’s only what happens if you combine Bush the Elder and Reagan. That’s not an entirely fair combination, as while they were both Republicans, they were very different kinds of Republicans. Reagan was a hawkish small-government conservative of the Goldwater tradition, while Bush the Elder was a moderate New England-style Rockefeller Republican. Reagan cut the top bracket’s marginal tax rates from 70% to 28%, while Bush the Elder raised taxes on the rich. To go by Silver’s description, you’d think it was Reagan who screwed the poor, and you’d expect Bush’s Clintonesque tax increase to have undone some of the damage. But that’s not what the data shows.
By Silver’s methodology (excluding transition years and measuring the overall growth from the second year of the administration to the last), here are how various income brackets fared under Reagan, Bush 41, and Clinton:
| 20th percentile | 40th percentile | 60th percentile | 80th percentile | 95th percentile | |
| Reagan | 10.62% | 9.87% | 12.06% | 13.47% | 16.36% |
| Bush 41 | -5.09% | -3.95% | -1.43% | -1.07% | -1.60% |
| Clinton | 16.16% | 13.97% | 13.24% | 13.24% | 15.08% |
To me, it looks like everyone did well under Reagan, while everyone did poorly under Bush the Elder, with the poorest being the hardest hit.
{ 6 comments }
I tend to be mistrustful of all of these comparisons between administations. They tend to look so agenda driven. How does one compare? Peak to peak? Trough to trough? Some sort of overall average?
Additionally, they tend to ignore things like consequences and elasticity. The effects of a tax levied in year X tend not to show up in year X but in years X+1, X+2, and so on and, oddly, those out years frequently land in different administrations.
Whatever the benefits of the Obama stimulus plan, its costs will be borne forever. That’s true of most policies.
Yup. Consequences of policy aren’t necessarily immediate. Over a four-year period, the business cycle swamps the effect of policy on the economy. These analyses completely ignore the effects of congess and the courts on public policy, and they ignore non-government events (technological changes, weather, supply shocks, etc) that can have very large effects on the economy.
My point was less to use this analysis to praise Reagan (although I do think Reagan deserves more credit for economic growth in the 80s than it’s currently fashionable to give him), but more to point out that changing a minor piece of methodology completely invalidated one of Silver’s major conclusions from the analysis.
The underlying percentile analysis doesn’t include any analysis of mobility. If you select your population groups by percentile at the start of the time frame and them compare the relative change in income, the slope goes the other way after a few years. The bottom 20% includes “starving college students”. The top 20% includes a lot of people nearing retirement.
True. And also don’t forget that these are household numbers, not individual income numbers, so changes in the number of working adults per household are being miscast as income changes (the long term trends are that the percentage of adults in the workforce in increasing as women continue to enter and stay in the workforce, but the average number of adults per household are declining as less of the population is getting married and staying married).
My point is that the underlying phenomena are too complex to harness to anybody’s partisan or ideological preferences. Just presenting the raw data is inherently misleading regardless of the interpretation.
Also, the data omits transition years — for example, 2001 is a transition year between Clinton and Bush, and it’s not clear who to credit/blame for the economy’s performance in that year, so I’m skipping it.
This tells me that his analysis is a complete crock, aside from the dishonest conflation of Reagan and Bush. Bush was president in 2001, but any economics 101 student can tell you that any change in monetary or fiscal policy takes 12-16 months for its effects to be fully realized in the economy.
If you really wanted to compare the presidents’ economic records, you would shift their terms by a year, so Reagan’s economic record would be evaluated from 1982-1989. Which changes things quite a bit.
Comments on this entry are closed.