What exactly happens in an economic recession, and how much has Katrina increased the likelihood of one developing?
Cafe Hayek, William Polley, and
Division of Labor all noted with dismay the statement from an economist quoted in a recent Washington Post story, suggesting that Hurricane Katrina could prove to be a source of economic stimulus:
“There will be a lot of rebuilding that is going to need to occur. These things do spur GDP growth,” said Ken Mayland, president of ClearView Economics in Pepper Pike, Ohio.
It is an implication of Keynesian models that in some circumstances, additional spending on any activity, such as trying to fix or replace all the things that are now broken, would boost the level of GDP. But even the most committed Keynesian would not attempt to extrapolate this principle to any and all situations, as if it were possible for Bangladesh to make itself ten times as rich as the United States if its government were only to invent enough make-shift jobs for its citizens to perform.
The position of any responsible economist would be that the level of a country’s GDP is determined by the interaction between what the nation is physically capable of producing, given the size and quality of its workforce, capital stock, raw materials, and technology, and whether the economy is employing those resources efficiently. During an economic recession, we see the unemployment rate spike up and capacity utilization measures plummet, suggesting to many of us that economic recessions have largely to do with that second issue of effectively utilizing the economy’s basic resources. A spending stimulus has more ability to boost GDP in such circumstances than it would when the economy is already near full employment.
The level of prosperity in an advanced economy results from a high degree of specialization of labor, capital, and most importantly, the institutions (shops and firms) that coordinate their activities. What happens in an economic recession is that the market’s success at coordinating this specialization breaks down. Firms and workers suddenly find that the activity for which they are uniquely suited is no longer in demand, and the degree of specialization makes it infeasible to redirect these productive inputs immediately to other sectors. These sudden demand shifts can occur, for example, if sharp energy price changes and the uncertainty accompanying them produce dramatic changes in consumers’ purchases of items such as cars. A theoretical model of how this can occur was developed in a paper of mine published in the Journal of Political Economy in 1988.
Reductions in economic activity from these sectors can themselves be the cause of downturns in other sectors, with the possibility of a complete collapse of coordination such as the U.S. experienced in the Great Depression, in which nobody can work in part because, well, nobody else is working.
In a typical hurricane, there is some destruction of productive physical capital. But precisely because the physical trauma is relatively local, there is little impact on the ability to coordinate economic activity. Many people end up doing pretty much what they were doing before on the other side of town, with the rest of the nation able to serve as a buffer of available resources with which the rebuilding can be carried out. Macroblog documents that previous major hurricanes had very little macroeconomic impact.
Is Katrina any different? Although the effects on energy prices originally looked to be enormous, it now appears that we are already past the worst, with oil and gas prices falling again today. I am likewise hopeful that New Orleans’ hobbled port will not produce the insurmountable problems with getting goods to market that some had originally feared.
On the other hand, New Orleans itself looks to me like a coordination problem on the scale of an entire metropolitan area. Some of the people affected will get back to working quickly. But many who worked in schools, stores, or hotels seem quite unlikely to be performing a similar function on the other side of the river any time soon. Part of the sadness in looking at the great city under water is the realization that much of the magic that allowed it to function together as a cohesive unit has been destroyed.
How big is that effect economically? The estimate from the Congressional Budget Office of 400,000 lost jobs has been widely repeated by the press. I have looked at the CBO report, but did not see exactly where the 400,000 figure came from. The CBO did call attention to the facts that about 620,000 people worked in the New Orleans-Metairie-Kenner Metropolitan Statistical Area last year, of whom 250,000 worked in the Orleans Parish. It does seem a safe guess that at least 200,000 of those previously working in the MSA will not be employed again for a period of several months at a minimum.
How big a shock is that relative to the national economy? For comparison, U.S. nonfarm employment fell by over 400,000 jobs in the first three months of each of the last two recessions, and would normally be expected to grow by significantly more than 400,000 jobs over a typical 3-month period. Thus by this very crude reckoning, Katrina by itself represents less than a quarter of the size of a typical recession-inducing shock, or less than half, if you use the CBO estimate of 400,000 lost jobs.
The other reason that Macroblog found very little evidence of an economic slowdown from other historical hurricanes is that usually hurricanes are completely unrelated to anything that may be happening in other parts of the economy. In the present case, I am concerned that, even without Katrina, we could easily see additional layoffs in autos and
airlines. And sooner or later, the degree of Fed tightening that we’ve already seen is going to exert a substantial contractionary effect on the economy.
So where does all that leave us? The situation economically is not as bad as it could have been, and not as good as we would wish. A recession in 2006 remains a possibility, but not a certainty. Statements like this one from Federal Reserve officials remind me that the Fed still has the power to turn it from the former into the latter.
I spent a lot of time trying to think about related issues while working on a paper about the possible costs of a massive computer worm incident, and I remain confused.
To try to illustrate my confusion, let me divide the workers of New Orleans into two classes which I’ll refer to as “essential” and “non-essential”. The essential folks are the ones so economically necessary that their employers will do what it takes to keep them working (the folks in the oil/gas industry, the electricity workers, the port workers, etc). They are working, and living in a tent outside New Orleans let’s say.
The non-essential workers are the haircutters, shop-clerks, etc who used to provide services to the essential workers. They are now out of a job, and living in Houston. A simplification, obviously, but perhaps illustrative.
Now, the essential workers used to spend money on products and services provided by the non-essential workers. Due to the physical disruption, they no longer can. So either, a) they spend that money on other services from other providers who they can get to, or b) they put it into the bank. If a), then the amount of economic transactions occurring has not been reduced, just redistributed. If b), the bank is going to take the money and lend it to someone else, promoting economic activity somewhere else (by increasing the supply of credit a little bit).
In the meantime, the government is borrowing a bunch of extra money to give to the non-economic workers who are now unemployed. So their spending power is not reduced as much as we might have feared (and not as much as the displacement of the spending of the essential workers has increased economic activity elsewhere). This reduces the supply of credit a little bit.
In short, has the level of economic activity really changed, or just shifted around?
Stuart, I find it easier to think about the real flows of goods and services rather than money. (We could do it in terms of money, but then we’d have to get into the question of how much those dollars buy). Your essential worker now has to drive an hour somewhere to get that haircut. From society’s point of view, an hour of her time and the gas for the drive have just been thrown away to deliver the same thing she used to get for fewer resources. She therefore can’t possibly consume the same quantity of things she likes (haircuts and leisure) as she used to.
The reason it has to work out that way is that not as many people are producing stuff (haircuts or whatever) as they used to. If the former haircutters are now sitting around unemployed, we’re just not able to deliver as many haircuts as we used to. So the hours that your essential worker puts in are not going to be worth as much in terms of the other stuff the worker buys as they used to. Idle resources mean a net economic loss.
And of course, the unemployed haircutter is much, much worse off in terms of the stuff he can buy. If you help him with unemployment insurance, the taxpayer is worse off. No need really to track down all these flows– the ultimate reality is, fewer people working means less stuff gets produced.
Thanks for a very useful post. As you say, what has happened in New Orleans post-hurricane is that there is now a coordination failure. The restaurants won’t start business because the movie halls won’t, and vice versa. It’s like a great depression, acted out on the scale of a city.
Will the city depopulate? In this case, the flexible labour markets elsewhere will absorb these individuals. And, the economy will have suffered two kinds of shocks. First, the depreciation of physical capital in New Orleans. And, the waste of human capital where humans have to learn new things and shift careers in the practical vicissitudes of the labour market.
As many people have pointed out, in the past, natural disasters of this scale didn’t matter much. Did these natural disasters take out a city, and hence, are they comparable? I also wondered: Over the years, the extent of interconnections and complex contracting that makes up a modern economy has gone up. Hence, the costs of a disruption should be higher, as time goes by.
JDH: Your JPE model seems to be one in which monetary policy would be ineffective (because prices are flexible). Obviously, however, you do believe that monetary policy is effective, and you seem to have some idea of how sticky prices might interact with the phenomena in your model to make monetary policy sometimes effective in preventing (or at least ameliorating) a recession similar to the kind that your model contemplates. On the other hand, if the real world is at all like the one in your model, I would expect that such antirecession policies would cause inflation, at least in those cases where they are pursued aggressively enough to succeed in preventing recessions. Are you taking the general position that this is a worthwhile tradeoff (i.e., that the Fed should sometimes be willing to pursue inflationary policies in order to prevent recessions)? Would you argue that the Fed’s credibility today is sufficiently high to prevent a permanent impact from such inflationary policies?
Very good point, knzn. I’m not suggesting that a monetary expansion could revitalize New Orleans, nor am I advocating a monetary expansion. I am arguing that a monetary contraction– a deliberate effort to produce a level of output that is below what the economy could efficiently produce– is under way, and that this has the potential to make a bad situation worse. I’m further arguing that there are some similarities and potential for mutual reinforcement in the propagation mechanisms for these “supply-side” and “demand-side” effects, though you are correct in pointing out that I have not produced a formal mathematical model in which this last claim is illustrated.
In my humble opinion, the New Orleans destruction can prove be more serious than you can think, not because it disrupts the gas production and distribution system, but because it disrupts the grain transport system.
The cheapest way to transport grains to the international market is use the Mississipi river to transport it. To that, the transport system need an oceanic port where the Mississipi find the sea to ship the grains to other countries. That port is New Orleans.
That is an important problem because there are other countires that compete with USA for the grain international market. Brazil is a good example. So, if the USA grain transportion system is disrupted or if USA need use more expensive ways to export the grain, USA will lost markets. The farmland will lost profit because the farmland cannot sell the grains to other countries and the farmland will be forced to sell the grains at lower prices to the USA domestic market. I see that an desincentive to the farmland produce, so it is possible that USA will have a lower farmland production if the grain transportation system stay disrupted for long time.
Now, ports need workers to operate them. Workers need homes and an economic infrastructure to support them (for example, the workers need buy food somewhere). That is a main factor that will determine how much time the NO destruction will affect the port return to full operation and how much time the grain transportation system will be affected.
I think that if the predictions that the New Orleans port return to fully operational only next year that the next year farmland production will be lower. The farmers will lost money this year because they will have no cheap way to transport the grains to the external markets and they will have less money to invest at the next year production.
So, my advice is stay tunned at how much the NO destruction is affecting the grain transport system. It can be more important to USA macreconomy than the oil system disruption (that aparently is solved or will be solved soon).
Joo Carlos
sorry the bad english, my native language is portuguese.
JDH: “a deliberate effort to produce a level of output that is below what the economy could efficiently produce” Do you really mean that the way it sounds? That perhaps the Fed, barely 2 years after a deflation scare, is back to the old game of keeping output below potential in order to get the inflation rate down (or maybe to stop the housing boom)? Or are you just saying that the Fed has made too low an estimate of potential output?
In evaluating the impact of prior hurricanes it is important to note significant variants. It is widely assumed that much of the damage will be uninsured flood losses vs the normal insured wind damage resulting in a greater loss of capital for the affected.
A second variant is that the coastal areas of the Carolinas, and Florida that have been most affected have had minimal impact on industrial output, except for citrus prices. While gasoline price spikes may be temporary, the impact on natural gas prices is likely to be a longer term issue that will impact the home heating cost of much of the country.
The affect on agricultural exports has been noted, but the affects on farm incomes will be compounded by other factors. Cheap domestic feed stock prices will result in heavier weights of livestock going to market likely depressing livestock prices, and fertilizer prices will probably increase significantly due to the cost of natural gas. The farm belt is in for a tough row.
Combine these new factors with other weak areas in the economy – notably airlines, auto manufacturing, and the possibility that home building has peaked and we have the makings for a contraction.
Bill
Great commentary. I have been milling with the dichotomy between the static costs and the dynamic, long-term impacts of Katrina. Your post clarifies some of my ideas.
Andrew
Joao and Bill, if the market expected the port situation to disrupt grain shipments significantly, I would have expected the price of grain in Chicago to have collapsed. But corn is unchanged and soybeans are up in price compared to where they were August 26.
knzn, neither I nor the Fed know exactly where potential output is. My concern is that the Fed intends to keep on raising interest rates until they see inflation actually coming down. The problem with that strategy is the delays in the effect of monetary policy– inflation won’t start to decline until after output growth has started to slow, and the output effects will come significantly after the interest rate hikes.
The indicators that can give the Fed immediate feedback are the rate of change of the fed funds rate over the last 12 months and the slope of the yield curve. As I’ve argued in previous posts, both of these suggest that the Fed has been moving pretty aggressively in a contractionary direction.
“Although the effects on energy prices originally looked to be enormous, it now appears that we are already past the worst, with oil and gas prices falling again today.”
NPR’s Morning Edition had a talking head claiming that, due to previously higher prices and the hurricane, residential natural gas will be 70% more expensive this winter. Expensive natural gas will also raise electricity prices. Although crude prices are down because the reduction in refining capacity has reduced demand for crude, distillate prices are still higher. I’d expect gas to stay expensive through winter, and heating oil prices to be high as well. Has the temporary loss of refinineries translated into a setback for reserve accumulation? Gonna be an expensive winter…
As for rebuilding – that’s activity that shouldn’t be included when calculating whether there’s a recession, unless there’s also a way to offset it against the premature destruction of assets and services from the hurricane. While much of the rebuilding will come through debt financing (federal and personal), lots of it is/will be coming through charity, savings, or insurance payouts. The pain will be spread around, but it will still hurt – and it isn’t close to being over yet.
JDH wrote:
“Joao and Bill, if the market expected the port situation to disrupt grain shipments significantly, I would have expected the price of grain in Chicago to have collapsed.”
Professor, the key words are “if the market expected”. I really don’t believe that the market have a crystall ball to see the future. Let’s say that dot com bubbles happens.
For now the New Orleans port is not operating. We don’t know how much time the port will be not operational. IMHO, it is possible that the port will be not fully operational for at least 6 months. If so, the farmland production that will start this fall will have problems with transport. Next year farmland will be affected if the profits next fall are affected because it is not possible to transport the grain to other countries at lower coust. If there a problems to produce (because the refineries are off) and transport (because the port is off) fertilizants to the farmland using the Mississippi the farmland problem will be worse.
However this year production will be not affected (the problem will be transport and not production) because the crops where sewed last spring and they will produce if there is transport or if there is not transport. But next year production can be affected if the farmlanders lost money this fall because they will have less money to invest because they cannot transport the grain to other countries to have a better selling price.
A good question is how much the INTERNATIONAL prices and the DOMESTIC prices will be affected. I think that the domestic USA prices will go down while the international prices will go up if the NO port continue to be not operational. Next year both prices will go up if domestic US farmland production get lower because the farmlanders have less money to invest.
Joo Carlos
sorry the bad english, my native language is protuguese.
Professor
The price for grains is rather dependent upon world markets and production. The social disruption occuring in Venzuela, a significant grain producer, is expected to have some market effect, and the drought in the midwest this year is likely to support higher prices. However, on the farm prices reflect storage costs and transportation costs, and can be quite disjointed from terminal prices.
Wall Street Journal 9/8/05 , page A10 – …. corn prices in central Iowa were offering $1.88 for January delivery versus $1.65 now.
The futures spread between September and December deliveries is $.17, while the rural spread is twice that. Those areas with strong production are likely to encounter significant storage problems and the traditional overflow is on the ground storage that results in significant waste.
Back to the WSJ article. The prices in Osceola Arkansas (on the Mississippi) are 23% lower for corn and 8% lower for soybeans than at the same time a year ago.
The economics of farming are not a new problem, and the interuption of exports due to Katrina may not be an overwhelming problem. However, low production, low field prices, and high fuel costs are going to hurt this fall. Add increasing interest costs and the prospect of higher fertilizer costs next year and it is not a very strong picture.
Yes, there are clearly disruptions to the local economy, the port, the oil sector, and the grain sector. However, an important reason we have almost never (maybe just plain never) seen a recession develop from a hurricane is that it offers an increase in aggregate demand related to the rebuilding. Heck, we will now have an approximately $60 billion fiscal stimulus from the federal government, not offset by tax hikes, and probably not offset by extra tightening by the Fed, which will help finance this reconstruction effort.
Recession? Maybe it will trigger a boom a few quarters from now.
September 2005, gas cost $4.50/mmBTU. Today, it’s down a bit at about $11.50. September is one of the seasonal price valleys – no winter heating and no summer electric peakers.
With 20% of the US electricity fueled with gas, assume that a third of the total generation cost is gas-fueled (coal and nuclear are cheaper and make up most of the balance). Generation costs are about half of retail price. (There is more to the calc. Email for more.)
Therefore, the nation’s electric bills will run about 10 to 15% higher from 18 months ago due to higher natural gas prices. Just about every business activity pays an electric bill so that will eventually appear as general price increases.
Let’s hope the Feds don’t try to wring this “inflation” out of the system with higher interest rates because it will be counterproductive. It will just raise the cost of alternatives to NG-fueled plants unless it destroys demand – people stop heating and lighting their homes and business reduce production.
Declining EROEI could produce an inflation that monetary policy won’t fix.
JDH: “The indicators that can give the Fed immediate feedback are the rate of change of the fed funds rate over the last 12 months and the slope of the yield curve” I think the Fed would argue that the level of the funds rate (relative to some reasonable measure of expected inflation) is more important than the rate of change (hence their “removing accommodation” language). I’ve seen some evidence that supports this view.
They don’t seem to have a good answer for the yield curve argument, though. “It’s a conundrum” basically means they think they are smarter than the bond market, but they aren’t exactly sure how. The only substantive explanation I’ve heard is that plentiful foreign capital is keeping long-term rates low, but I have trouble with that one. Wouldn’t foreign investors, given the flat yield curve, prefer the safety of short-term investments? And shouldn’t this foreign demand, in the face of rising short-term rates, therefore be reflected in the value of the dollar rather than the slope of the yield curve? (Anyone out there have a better idea?)
Joseph: “Declining EROEI could produce an inflation that monetary policy won’t fix.” Monetary policy can fix it by producing a recession (or a depression), so that the price of everything else goes down to offset the price of fuel. And maybe I just answered my own question from the last paragraph. If the Fed is really worried about the inflationary impact of fuel costs, and if their beliefs about the impact match yours, they may be deliberately trying to cause a recession, which would explain the shape of the yield curve.
The conumdrum of the inverted (or nearly inverted) yield curve is reflective of investor psychology. We expect that what happened yesterday will happen again tomorrow. For the prudent investor we had the end of the tech stock boom, and extremely low C D rates within the last five years, and thus many individual investors expect that equity investments will not do well, and that C D rates will remain low. Thus, they are attracted to the ten year Treasury as a refuge.
Basic investment theory requires an evaluation of risk. Fear is simply risk to the Nth power.
Bill
Bill: If people are concerned about risk, then why are credit spreads so narrow?
JDH:
If we think of it in terms of the physical economy, then it’s still not obvious to me. Sure, the productive capacity of much of New Orleans has been eliminated. But there could very well be that much spare capacity lying around the economy (New Orleans was not a very big city). If aggregate demand didn’t change, then people who would have bought products and services from residents of New Orleans can go get something else from somebody else instead. Only if there is no spare capacity elsewhere, or no spare capacity for critical goods (eg oil/gas, or mouth-of-Missisipi port services), does that seem to me to be guaranteed to lead to a reduction in production. But my suspicion is the effect on oil production is not going to be large enough to cause major economic changes – the port problems may be the largest supply-side issue.
Thus it almost seems to me that the main effect of this kind of thing actually is the psychological one on demand (which seems very hard to predict – it has somewhat of a tendency to promote borrow-and-rebuild (increasing aggregate demand), but maybe has something of a chilling effect on the national mood and thus overall consumer confidence.
(This is why I have made very little satisfactory progress on projecting the economic impact of computer worms – there too I couldn’t convince myself that the largest effect wasn’t psychological).
Chart of the Week: DJIA Making Higher Lows
The DOW has made higher lows and is now in the process of pulling back to its trend line. This is consistent with a minor retracement from modestly overbought levels. DJIA Making Higher Lows Source: Stockcharts.com Note that if Traders get a sense that…
I have always looked at stocks as the primary leading indicator, determinate of grain prices.
Historically, grain stocks shoot up when world stocks are drawn down sharply.
The closing of the NO port will impact shipments of grains for a time, but it will have essentialy no impact on world grain stocks –although for a short period stocks will not be distributed optimally — so I would not expect it to have a significant impact on grain prices.
I think the real danger of Katrina is not the oil or local disruption but the huge federal bailout, the reduction of the Fed government to first responder, implications of more in the future politically, the East Germanization of the poor welfare folks from LA, etc.