The first of more than $110bn in tax rebates for US households start arriving on Monday, amid a vigorous debate as to how big an impact this emergency stimulus – and accompanying investment incentives – will have on the US economy.
Some analysts, including Macroeconomic Advisers and Goldman Sachs, believe it could deliver as much as a 2 percentage-point boost to annualised growth in the quarter in which it kicks in.
Others believe that the effect will be much smaller. Bank of America sees a growth boost of a bit under 1 percentage point, Merrill Lynch as little as 0.5 of a percentage point.
The difference depends on how much of the rebate package will be spent and how much will go on imported goods. It is also related to the time-frame over which it is spent, and whether this expenditure will, in turn, trigger knock-on spending.
In addition, there is a technical dispute over how to express the boost in terms of annualised quarterly growth. On this point, the rebate bulls look to be right, while others may be underestimating the initial impact of the stimulus on quarterly annualised growth by a factor of four.
Maths mistakes |
Many top Wall Street groups are underestimating the short-term impact of the tax rebates on growth by a factor of four, owing to a maths mistake. To understand why, consider the effect of a $20bn boost to spending on the annualised growth rate of the quarter in which it takes place. Most analysts did their sums like this: $20bn annualised is about $80bn. That is 0.8 per cent of annual consumption. Since consumption constitutes about 70 per cent of gross domestic product, that equates to slightly less than a 0.6 per cent boost to annualised GDP growth in the quarter. But the Bureau of Economic Analysis - which compiles the GDP series - says the maths here are wrong. The correct way to do the sum is as follows: the level of consumption goes up by $20bn or 0.8 per cent in the quarter. But you then have to annualise this increase to get an annualised growth rate, which works out at 3.2 per cent. Since consumption constitutes about 70 per cent of GDP that equates to about 2.2 per cent extra growth - a four times bigger boost to annualised growth in that quarter than the incorrect approach suggested. Of course, the actual extra amount spent in the quarter remains the same - $20bn - only the reported growth rates are different. |
In effect the government will nationalise part of US household debt – socialising some of the costs of the economic downturn.
In doing so it may reduce the risk of a sudden pull-back in spending by overstretched consumers, even if it does not actually boost spending by much.
Analysts estimate anywhere from 20 per cent to 50 per cent of the rebates will be spent over a period of four to six months.
The Treasury has rushed to get the rebates out ahead of schedule, in the hope of arresting a deepening economic slide.
Hank Paulson, Treasury secretary, now says that $50bn (€32bn, £25.2bn) will be distributed by the end of May, with the first payments sent by electronic transfer this week.
This means the effect may first be seen in May and June retail sales. Some analysts believe the impact will be primarily on second- and third-quarter growth, others on third- and fourth-quarter growth.
The Congressional Budget Office admits that there is “some uncertainty about the rebate’s effect”.
Theory suggests that since households generally base their spending decisions on expected lifetime income, a temporary boost to income should induce spending only by those who are “liquidity constrained” – unable to access cash.
Studies of the 1975 rebate suggest that only about 12-24 per cent of the package was spent in the quarter in which it was received.
However, analysis of the 2001 tax rebate suggests that 20-40 per cent was spent in the first quarter it was received, and about two-thirds over two quarters.
There are, however, important differences between the 2001 rebate and this year’s package. The 2008 rebate is much bigger – $170bn over two years, with $110bn in the first year for consumers.
The amount per household is roughly double – $600 per individual, $1,200 for a couple and up to $600 extra for each child.
Moreover, the current rebate is more skewed towards lower income workers than the 2001 package. Since lower income workers are more likely to be cash-constrained, theory suggests that should mean it has a bigger effect on spending.
Household circumstances have also changed since 2001. They are more indebted – suggesting they may save more. However, a greater number may also be cash constrained, which suggests that they might save less.
In one crucial respect the 2001 and 2008 rebates are fundamentally different. The 2001 rebate was a downpayment on a permanent tax cut; the 2008 rebate is not. This suggests it may have a considerably smaller impact on spending.
Whatever the size of the boost to spending, the effect will be temporary – six months at most – and there will be a negative effect on reported growth when it fades.
Policymakers hope it will mitigate the risk of a recessionary dynamic taking hold in the meantime. In doing so, it could provide a bridge to what they hope will be an economic recovery beginning some time in the second half, as residential construction stabilises and companies complete inventory adjustments.
But if the underlying conditions do not improve during the half year of stimulus, the economy could relapse into recession as soon as it expires.
