Don’t look at the rich?
My last post, arguing that the share of US income going to the top 1 per cent of households is now so great that any effective policy must be financed by reducing or more effectively taxing the income of this group produced a range of interesting (and some not so interesting responses). First up, it elicited what appears to be new variants on a couple of standard rightwing talking points. More interesting to me is a response from Matt Yglesias arguing (as I read him) that, even if there is no serious prospect of reversing the shift of income to the top 1 per cent, there is still plenty of capacity for progressive political actions based on a broadly neoliberal (US sense) agenda.
The new talking point (apparently due to the Heritage Foundation) starts with the observation that while US tax policy has changed radically since the 1950s (particularly with the elimination of high marginal rates from the income tax system), the share of national income collected as revenue by the Federal government has not. This, it is claimed, shows that tax rates are uncorrelated with revenue. The claim, then, is a watered down version of the Laffer hypothesis. Whereas Laffer claimed that the US (as of the late 1970s) was on the descending portion of the Khaldun curve, the new talking point suggests that the US has, for the last 50 or 60 years, been on or near the flat part at the top. I haven’t seen the claim supported by any analysis the mix of tax revenues between income taxes and other taxes, and I don’t intend to offer one. I’ll just make the points that
(a) most other developed countries have higher tax rates than the US and collect a larger share of national income in tax revenue than the US
(b) numerous studies of large tax changes such as the Bush tax cuts have confirmed the common-sense expectation that higher tax rates produce more revenue
The Bush senior assessment of Laffer’s work as ‘voodoo economics’ remains as valid now as when he first made it.
The second talking point is that we shouldn’t worry about income inequality, since consumption is what matters. In the absence of any actual data, a lot of the discussion is based on the assumption that the income of the top 1 per cent is not consumed, but saved and invested, and that if the government taxed these investment flows it would have to replace them with more public investment, leaving no revenue to finance transfer payments or public.
Since there is not a lot of data on the consumption patterns of the top percentile of households, it’s unsurprising to see argument based on anecdotes, in this case with people like Bill Gates in mind. It’s amusing though to note that very different anecdotes were being cited only a few months ago when the Bush tax cuts were in danger of expiring. At that time, we were regaled with household budgets showing that those on $250k could barely make ends meet after covering the basic expenses associated with their lifestyle. Admittedly, these budgets usually included some pretty hefty 401(k) contributions, but there was also plenty of consumption expenditure.
It is certainly true that, on standard measures, consumption inequality is lower than income inequality. In part this reflects the fact that a significant component of income inequality reflects transitory variations, which households can smooth over by borrowing and saving.
Although consumption inequality is than income inequality, we would expect the two to move in parallel. In the US, standard measures suggest that income inequality has grown much more than consumption inequality. Some recent research by Aguiar and Bils (full paper paywalled) suggests that the standard measures are misleading in various respects, such as underestimating growth in the luxury expenditure of the wealthy.
Taking the figures at face value though, the big factor driving a divergence between measured income and consumption inequality over the period from the 1970s to the recent past was the increase in household indebtedness over this period. According to the standard measures this enabled middle-income Americans to increase consumption substantially even though median household income was growing only slowly (and has actually dropped in the last decade).
Such a trend couldn’t be sustained and it hasn’t been. As credit has been tightened, consumption has weakened in general, but particularly for credit-constrained households faced with declining incomes. It seems pretty clear now that consumption inequality is growing.
I’ll turn now to the more interesting response from Matt Yglesias
the median American household has quite a lot of money compared to the median household of almost every other country. And yet, I think there are a lot of other respects in which quality of life in the United States falls short. We spend a lot of time in traffic jams. We have both a frighteningly high murder rate and a frighteningly high level of incarceration. Our health care system is very inefficient. Americans work very long hours and have unusually little vacation time. It’s not clear to me that any of these issues can be usefully tackled primarily by focusing on higher taxation of the very wealthy.
Looking at this list, it’s probably true that that the high murder rate is due to a gun culture that predates the rise of inequality. But traffic jams are, to a large extent, a reflection of inadequate infrastructure and poor public transport. Fixing those things costs money, and the point of my post was that the only place to get that money was out of the 25 per cent of US income going to the top percentile of households.
The same is true of health care: any really effective reform would require more public expenditure, at least in the short run. But high inequality also contributes to US health costs more directly – a substantial part of the cost gap with other countries is due to the high incomes of medical professionals, which in turn reflect both the general inequality of the income structure and the huge costs of medical training, also due to inequality
As regards working hours, the end of the long decline in annual working hours for US workers coincides with the emergence of wage stagnation and growing inequality. Households have supplied more labor in an attempt to maintain living standards. This point has been made for at least 20 years (I can’t find it online, but the first time I saw it in detail was in a Congressional Study: Families on a Treadmill: Work and Income in the 80’s, January, 17, 1992)
fn1. I don’t want to misrepresent Matt. He certainly endorses more progressive taxation and I assume he thinks of the ideas he’s putting forward as complements rather than substitutes for redistribution. But the post still suggests that it’s a waste of time to put too much energy into a fight that’s doomed from the start. My view, on the contrary, is that a challenge to the top 1 per cent is the only political strategy that is likely to produce an effective mass mobilisation, though I admit I have no idea how to get started on this.
fn2. As Laffer himself concedes, the correct idea behind the curve associated with his name was well known as far back as Ibn Khaldun in the 15th century and were certainly familiar to Keynes. Laffer can reasonably claim originality for what I’ve called the Laffer hypothesis, that the US is, or was, on the declining part of the curve. So, as in other cases, Laffer’s napkin presentation contains much that is correct and much that is original. Unfortunately …