« Bargaining with your roommates | Main | Robin Hanson and Will Wilkinson »
Stoking the fires
Ken Rogoff writes:
I am puzzled that so many economic pundits seem to think that the solution is for all governments, rich and poor, to pass out even more cheques and subsidies so as to keep the boom going. Keynesian stimulus policies might help ease the pain a bit for individual countries acting in isolation. But if every country tries to stimulate consumption at the same time, it won’t work. A general rise in global demand will simply spill over into higher commodity prices, with little helpful effect on consumption. Isn’t this obvious? Yes, there is still a financial crisis in the US, but stoking inflation is an incredibly unfair and inefficient way to deal with it.
In other words, if the initial shocks are real, boosting aggregate demand won't have much of a positive effect plus it will worsen inflation. Maybe it's a good thing the stimulus package is too small to be very "effective." Mark Thoma adds comment. I am, however, puzzled by Rogoff's distinction between a single-country stimulus package and the global combination of such packages. If nominal stickiness is a binding problem in enough sectors, a large enough stimulus can work no matter how many countries do it. If nominal stickiness is not a major binding problem (and I suspect this is the relevant case), then even a single-country stimulus plan will be ineffective.
Posted by Tyler Cowen on July 7, 2008 at 07:23 AM in Economics | Permalink
Comments
His reasoning seems to be that an individual country can consume more without having a large effect on commodity prices, but if all countries spend more together, commodity prices will rise to choke off the attempted increase in demand. Its a strange argument. I've never heard an economist make the argument that commodity prices are an important macroeconomic stabilizer. Typically, you think of interest rates as playing that role. Since the debates over the Pigou effect, the consensus has been that goods prices really don't matter much as stabilizers. I'm not sure why commodity prices would be any different.
BTW, Tyler, I think you're wrong about nominal stickiness being an important consideration here. Prices can be flexible and rise and the economy will remain above potential as long as the poicy interest rate is too low. Remember, if the problem is too much demand overall, then all prices should rise together, more or less proportionally, and there should be very little real effect on demand.
Posted by: steve at Jul 7, 2008 9:46:49 AM
Since the debates over the Pigou effect, the consensus has been that goods prices really don't matter much as stabilizers. I'm not sure why commodity prices would be any different.
These theories date from an era where currencies were (and almost always had been) fixed to a commodity standard (usually gold). Long-term inflation expectations were almost entirely absent from the data these economists studied, because the supply and demand of most commodities rarely ever made a radical long-term departure from the price of precious metals. We live in a very different world now, where currencies float independently of commodities, and the long-term relationship between them may become arbitrarily large.
Consider what happens during a severe credit crunch where central banks inflate the supply of floating currencies, relative to demand for them, to add liquidity to try to stop the crunch:
(1) there is a flight to safety, causing interest rates on safe debt (e.g. U.S. Treasuries) to drop.
(2) arbitrage between artificially low central bank rates and rates for safe debt also cause rates for safe debt to drop (and central banks sometimes perform this directly through open market operations -- and recently the Federal Reserve has even been buying up large amounts of unsafe debt).
(3) central bank reaction to lower interest rates and otherwise add liquidity causes an increase in long-term inflation expectations, causing long-term interest rates and commodity prices to rise.
The effects of (1)-(3) on interest rates for long-term safe debt can cancel each other out, while commodity prices can rise dramatically. Under floating currencies, the ideal safe portfolio is a mix of government bonds and commodities. The two are anti-correlative: if the money supply inflates, commodities rise and bonds fall, and vice versa. The idea that Treasuries alone are a safe investment is an obsolete idea left over from the Bretton Woods era, and it should be no surprise that we see retirement funds, sovereign wealth funds, and many other long-term investors expanding their long-term long commodity positions to hedge the debt in their portfolios.
Long-term inflation expectations hardly ever increased appreciably under Bretton Woods and previous precious metal based standards: indeed then deflation was the more common problem during credit crunches as money was hoarded. These short-term inflations and deflations tended to correlate with with cycles of credit boom and bust and largely balanced each other out over the long run. Today inflation caused by central bank reaction is the main consequence of a credit crunch and is hardly ever reversed in the long run. Instead long-run inflation accumulates. Even where central banks eased during a credit crunch or to fund a large war, gold standards and similar were usually a credible commitment by money issuers to reverse the expanded money supply as soon as the credit crunch or war was over. Today's investors are acting as if under floating currencies this credible commitment no longer exists.
As I've shown before, it only takes a small change in long-term inflation expectations to produce a very broad-based commodity boom of the kind we've seen over the last decade and especially in the last year. We've had a great swing in inflation expectations from the late 1990s, when it briefly appeared that deflation might be as common as inflation under a floating currency, to today, when many believe the 1970s may be more typical of floating currency behavior. Thus the great increase in commodity prices from 1998 to last year. Add a credit crunch, and central banks' inflationary reactions to the credit crunch, and we see the effects on commodities and bonds that we've seen this year.
Posted by: nick at Jul 8, 2008 3:40:39 PM
If nominal stickiness is a binding problem in enough sectors, a large enough stimulus can work no matter how many countries do it. If nominal stickiness is not a major binding problem (and I suspect this is the relevant case), then even a single-country stimulus plan will be ineffective.
Erm... No. Oil is a tradable, so a single country stimulus plan would be effective (at least partially). Even if the economy is large, like the US is, its demand is still just part of worldwide demand for oil. However, as Rogoff points out, if all countries attempt to stimulate demand then the spill into commodity prices would be larger, especially considering the relative low elasticity of oil demand.
Posted by: Nc at Jul 11, 2008 4:23:21 AM
Mabinogi online gold
Mabinogi money
Mabinogi Gold
cheap Mabinogi gold
buy Mabinogi gold
2moons dil
2moons gold
buy 2moons dil
2moon dil
cheap 2moons dil
Flyff gold
flyff Penya
flyff money
buy flyff penya
cheap flyff penya
cheap flyff gold
aion gold
buy aion gold
cheap aion money
aion money
cheap aion gold
Dofus kamas
buy dofus kamas
cheap kamas
dofus kama
dofus gold
dofus money
Knight Online Gold
Knight Gold
Knight Noah
Knight Online Noah
Posted by: aion at Jul 8, 2009 9:39:38 PM
