In the essay below I begin to draw on history to help understand the current fiscal stimulus. I will look at Robert Barrow's claim that WW2 multipliers have implications for modern day multipliers- and will attempt to prove that there is no linkage and that one must be wary of using history in the wrong context. In later posts I will look at more relevant examples of history, in particular the Japanese 'Lost decade'.
Keynes first coined the Multiplier concept in his theories on employment, interest and money. Given that it is a fiscal stimulus issue, over the last few decades, with Monetary policy leading the battlefront, no consensus estimate of the multiplier has been reached. This is largely due to the ‘no one size fits all’ complex, but is also due to a huge range of multiplier estimates from 0 to 3, with current Romer-Bernstein estimates of 1.5 (For increase in G) being a middle-ground.
The main problem is in fact that in economics, it is much harder to pursue ‘laboratory’ based evidence as in the harder sciences. Thus it accounts for why Romer-Romer estimates of a tax multiplier of 3 does not apply in reality- a lot of the estimates of the multiplier are exogenous/ceteris paribus estimates, which end up simplifying the economy, and do not take into account changing expectations on the part of consumers and other endogenous variables that may be affecting the Government spending increase/tax cut’s affect on the economy. Given that GDP is a function of so many variables, and given that expectations play such a significant role in determining those variables, it is extremely hard to pin-point the effect of a tax cut or spending increase simply because it is hard to predict how expectations will adjust…an example I have dwelled on previously is Ricardian Equivalence.
Take one such example, World War 2. Estimates I have seen are 1 by Valerie Ramey and 0.8 by Robert Barro. Given the enormous size of the fiscal stimulus (roughly 5.2 Trillion in comparative terms today), even a 0.8 multiplier meant that it did accumulate to a large increase in GDP. Krugman, including others, believe that the return of monetary policy combined with WW2 ended the Great depression, and enabled the post WW2 boom.
Now, do multiplier estimates in the context of war and peace really matter now? Robert Barro says that these are more reliable simply because it is much harder to calculate multiplier effects in light of business cycle fluctuations, relating to the endogenity problem discussed in the multiplier. But there are many good reasons why multipliers from WW2 have no relevance right now.
Keynes first coined the Multiplier concept in his theories on employment, interest and money. Given that it is a fiscal stimulus issue, over the last few decades, with Monetary policy leading the battlefront, no consensus estimate of the multiplier has been reached. This is largely due to the ‘no one size fits all’ complex, but is also due to a huge range of multiplier estimates from 0 to 3, with current Romer-Bernstein estimates of 1.5 (For increase in G) being a middle-ground.
The main problem is in fact that in economics, it is much harder to pursue ‘laboratory’ based evidence as in the harder sciences. Thus it accounts for why Romer-Romer estimates of a tax multiplier of 3 does not apply in reality- a lot of the estimates of the multiplier are exogenous/ceteris paribus estimates, which end up simplifying the economy, and do not take into account changing expectations on the part of consumers and other endogenous variables that may be affecting the Government spending increase/tax cut’s affect on the economy. Given that GDP is a function of so many variables, and given that expectations play such a significant role in determining those variables, it is extremely hard to pin-point the effect of a tax cut or spending increase simply because it is hard to predict how expectations will adjust…an example I have dwelled on previously is Ricardian Equivalence.
Take one such example, World War 2. Estimates I have seen are 1 by Valerie Ramey and 0.8 by Robert Barro. Given the enormous size of the fiscal stimulus (roughly 5.2 Trillion in comparative terms today), even a 0.8 multiplier meant that it did accumulate to a large increase in GDP. Krugman, including others, believe that the return of monetary policy combined with WW2 ended the Great depression, and enabled the post WW2 boom.
Now, do multiplier estimates in the context of war and peace really matter now? Robert Barro says that these are more reliable simply because it is much harder to calculate multiplier effects in light of business cycle fluctuations, relating to the endogenity problem discussed in the multiplier. But there are many good reasons why multipliers from WW2 have no relevance right now.
i) The economy in WW2 was in a ‘fully employed’ mode, the war effort required the full use of resources, and increased labour effort in light of the extreme circumstance.
ii)A corollary from i) is that, given the large amount of resources employed, and the absolute size of the package, it is plausible to say that the multiplier suffered from diminishing returns. This is a simple fact, and can be illustrated by this conceptual graph:
O- Optimum E-estimateWhat the graph shows is that the fiscal stimulus of WW2 was likely to be past the optimal level of Government stimulus. Thus it is likely that given the optimum government spending is the spending required to reach full employment, the excess government spending ended up causing inflation, reducing the multiplier to 0.8. Given that the current fiscal stimulus package is insufficient to fill in the deflationary gap, we are suffering from a deficiency of spending! Barro’s use of WW2 is not wise in light of this fundamental difference.
iii) Another good point is that, due to the war effort, there was a significant rationing of consumption and investment to allocate more resources to the war effort. This is true of most wars, and one can say that in a war effort there is a natural crowding out of investment and consumption, not necessarily via increased military spending, but due to the nature of war and the need to divert resources to warfare for the good of the nation. At the moment, there is no rationing as such, except individual rationing on the part of increased precautionary savings and an attempt to reduce the debt by households and firms. In any case, Barro has not established whether WW2 spending had any direct crowding out effects due to the rationing process, and hence it cannot provide a precedent for predicting the amount of crowding out for this fiscal stimulus. Krugman rebuts Barro’s analysis by showing declines in both spending on new homes and autos dropped considerably as part of the rationing process, and picked up steadily after WW2 (as balance sheets became healthy due to the reduced private sector debt to GDP).
iv) On the other hand, In recessions due to a financial crisis, which dampen consumption and investment considerably, the adverse balance sheet effects mean that investment is constrained until debt to GDP levels go to fundamental levels, in other words, deleveraging will continue until firms are recapitalised and ready to borrow again. Whereas consumption and investment were crowded out in WW2 via rationing, in the present day private sector borrowing is deficient purely because of insufficient consumption demand and balance sheet problems. Thus fiscal stimulus, as I have explained in previous posts, does not crowd out private sector investment!
The above reasons show clearly that it is unwise to use WW2 as a precedent for analysis of the current situation. Another problem of course is that in recessions, it takes time for investment and consumption to return to normal levels. This is simply because investment is based on profitability, and if current sales in a business are low, a business is unlikely to invest, even if they expect the economy to be gaining traction. In fact, given that the capital stock is reducing due to insufficient investment, investment will eventually increase due to the ‘use, decay and obsolescence’ of capital. All of this takes time, of course, and there is no doubt that whatever stimulus the Government provides will filter its way through to increased sales for businesses due to the multiplier effect, and can stimulate business investment in this manner.
Barro does have some arguments however. For example,
There are reasons to believe that the war-based multiplier of 0.8 substantially overstates the multiplier that applies to peacetime government purchases. For one thing, people would expect the added wartime outlays to be partly temporary (so that consumer demand would not fall a lot). Second, the use of the military draft in wartime has a direct, coercive effect on total employment. Finally, the U.S. economy was already growing rapidly after 1933 (aside from the 1938 recession), and it is probably unfair to ascribe all of the rapid GDP growth from 1941 to 1945 to the added military outlays. In any event, when I attempted to estimate directly the multiplier associated with peacetime government purchases, I got a number insignificantly different from zero.
The arguments above do not make sense at all. Though I agree with his first point in that wartime outlays and the fiscal deficits were temporary, meaning that the ricardian equivalence effect was constrained, the consumer demand was indeed offset- due to the rationing required to generate funds for the war effort.
For the second point, the fact that it had coercive effects on total employment meant that it became an over-inflated economy, largely different from the under-employed economy of today.
The third point overestimates the multiplier for wartime government purchases, not peacetime. Given that we are facing a prolonged slump, any multiplier estimates need not consider the rapid GDP growth due to the private sector- and it makes the exogenous analysis much easier, when keeping a ceteris-paribus assumption. Another problem of course is that Barro implicitly neglects knowing the multiplier he learnt in ECON 101. The problem of isolating investment growth from Government spending is that the multiplier attempts to show that increases in Government spending lead to increases in consumption and investment- via the increased income it initiates. This effect is shown in the IS-LM model, where this fact is even more pronounced in light of accomodative monetary policy, which ensures that both investment and consumption increase. I am of the belief that the use of WW2 to calculate current-day multipliers does not make any sense at all.
What I dislike about Barro's analysis is that fundamentally an analysis of fiscal stimulus using historical examples should focus on fiscal stimulus in a financial crisis. I feel Barro's argument is a classic case of a right-wing economist with arguments tailored to his side of the political spectrum, rather than arguments with a strong factual and analytical basis. At current, there are reasons to believe that the multiplier for current stimulus is greater than the WW2 multiplier.
No comments:
Post a Comment