Stimulating, selectively and statistically significantly | FT Alphaville

Stimulating, selectively and statistically significantly

The tax cut package is proving stimulative already – at least in terms of encouraging debate in the wonkosphere.

We continue to share the unoriginal view that extending the tax cuts for those earning above $250k will not provide the best bang to buck ratio. (And, yes, we accept that there are other reasons why you might want to offer tax cuts.)

Ezra Klein’s updated cascading blobs show that the average per taxpayer cut is – unsurprisingly – skewed toward the top end:

This matters because, as the Congressional Budget Office and others have pointed out, on average those further down the income scale will spend a higher proportion of their stimulus dollars.

The Economist’s Democracy in America cites this argument in an interesting post on Thursday. It riffs off a recent paper by Jonathan A. Parker, Nicholas S. Souleles, David S. Johnson and Robert McClellan, which looked at consumer spending in response to the 2008 stimulus payments.

While accepting that lower income people are more likely to spend a higher proportion of their income, the post also suggests that high income folk may, after all, be worth stimulating:

What’s interesting, though, is that total spending by both low-income and high-income recipients increased more than total spending by average recipients. In other words, though it’s not a statistically significant result, both poor people and rich people seem to have been more likely to spend their stimulus checks than average people were. The authors suggest that the main factor in propensity to spend may be liquidity: results also showed that homeowners were less likely to spend their checks than renters. One might hypothesise that you’re most likely to spend stimulus money if you’re really poor, but if you’re sitting pretty you’re still more likely to spend it than if you’re underwater on your house or saddled with credit-card debt. This would suggest that from a stimulus point of view, the people we really ought to cut out of the tax-cut extension are not so much the wealthy as the middle-class. Tax cuts for income under $35,000 and over $75,000, but not in between. “So because you are lukewarm, and neither hot nor cold, I will spit you out.” But somehow that doesn’t sound iike a political winner, and anyway there are some social-justice concerns involved there.

Rrright. The key bit here is “though it’s not a statistically significant result”. We therefore can’t say with (95 per cent) confidence that middle income people are less likely to spend.

As the authors write in the original paper:

As for results that further inform theories of credit markets and consumer behavior, across households, the responses are largest for older and low-income households, groups which have substantial and statistically significant spending responses. According to the point estimates, the responses are largest for high-asset households but this spending response is not statistically significantly different from zero.

Nevertheless, there might be something in this line of argument – albeit indirectly, via the tax cuts’ impact on the recent S&P rally and its corresponding wealth effect. Certainly, this is what analysts at Citi are arguing in this chart, released on December 7:

We’re too ill-informed to comment further agnostic on the argument at this juncture, so comments are very welcome.

However, following the logic of the Parker et al report (and the Economist post), it’s worth asking whether some commentators have been taking the easy way out in opposing the extension only for those $250k and above – should the threshold really be that high (if all one cares about is stimulus – which you might not)?

The paper continued a few other tidbits worth highlighting quickly. First, a defence of stimulus spending, per se, against Ricardian equivalence:

Although our findings do not depend on any particular theoretical model, the response is inconsistent with both Ricardian equivalence, which implies no spending response, and with the canonical life-cycle/permanent income hypothesis (LCPIH), which implies that households should consume at most the annuitized value of a transitory increase in income like that induced by the one-time stimulus payments.

Second, some food for thought for behavioural economists:

The estimated spending responses are statistically and economically similar for ESPs received by EFT (electronic fund transfers) compared to those received by mail, although there is little temporal variation in the former group with which to identify the key effect.

NB – This change in delivery method was supposed to nudge people into spending more than they would if they were handed a check, which was allegedly more likely to be put straight into a savings account.

And third, evidence on what people spent their stimulus money on:

The point estimates of the fraction of the ESPs spent suggest that, relative to the effects of the 2001 tax rebates estimated in Johnson, Parker, and Souleles (2006) (JPS), in 2008 the spending effect was slightly smaller for nondurable expenditures but more targeted towards durables. While this finding may be due to sampling error, it may also reflect some of the differences in the details of the tax cut and economic environment in 2008 compared to earlier

Which may go a bit of the way to explaining this unprecendented dislocation (chart from Maria Firorini Ramirez):

We still reckon ‘all’ is better than ‘nothing’ on the tax cut deal, especially since it includes a payroll tax holiday, which is perhaps the most direct stimulative tax measure available at a time when the labour market is still stuttering. However, we are sympathetic to those such as James Kwak and Simon Johnson who both eloquently warn of the fiscal dangers.

The Obama Administration’s defence is contained in this mini-lecture by Austen Goolsbee, Chairman of the Council of Economic Advisers, on the White House blog. He argues that there will be “no impact on the long run deficit of the country”.

Hmmm. We’re looking forward to the 2012 electorate versus bond market face-off on that one.

In the meantime, we’re continuing to weigh up how to put our extra dollars to good use.

Related links:
Taxes in wonderland – FT Alphaville
Looking down on the tax cut deal – FT Alphaville