September 19, 2005
Consumer confidence plunges
Yet another key leading indicator turns gloomy. How much can the stock market and the Fed shrug off?
The index of consumer sentiment produced by the University of Michigan is another closely watched barometer of where the economy might be headed next. The graph at the right (data from St. Louis FRED) displays the historical tendency of this indicator to plummet prior to or in the early stages of an economic recession (indicated by shaded areas on the graph). It's not good news that it fell 14% in the most recent month, bigger than the 11% drop in September 2001, and bringing the index to its lowest value in 13 years. The less widely followed Investors Business Daily and TechnoMetrica Market Intelligence index of economic optimism likewise experienced a 19% drop in September.
Such a response to Katrina has been the primary concern of many economists. One analyst dismissed the September numbers on the grounds that, "the index was likely weighed down heavily by the emotional response to the devastating images that resulted from Hurricane Katrina," to which Barry Ritholtz responded that summarizing the emotional response of consumers to events like Katrina is exactly what the index is designed to do.
Possibly what such positive spin envisions is that, as the recovery from Katrina and temporary nature of the disruption become clear, everyone will brighten up. But then again, here comes Rita, threatening to blast its way through the 44% of Gulf of Mexico crude oil production that is not still shut-in from Katrina, and toward the vital ports and refineries around Houston.
Anybody worried about that? Well, trading on NYMEX today pushed October oil up 7% and October natural gas and gasoline both up 14%. The latter figure amounts to 26 cents a gallon-- not bad for a day's work. Of the 40 cents a gallon decrease in retail gasoline prices that I promised here and here, we've seen 15 cents so far, but today's two bits could take a good bite out of what you've got left. Ah well, the NYMEX giveth, and the NYMEX taketh away.
Maybe those futures traders are just having an emotional response to this whole hurricane thing.
Posted by James Hamilton at September 19, 2005 02:24 PMdigg this | reddit
Listed below are links to weblogs that reference Consumer confidence plunges:
» Tying Together the Consumer Confidence Drop and Focus-Group Follies from BizzyBlog.com
The Mainstream Business Press’s attempt to “talk down the economy” is starting to work. Econbrowser has the grim news on consumer confidence (UofM site requires e-mail addy and name). I meant to post this comment at his site, but H... [Read More]
Tracked on September 20, 2005 02:30 PM
There is the constant call here for the Fed to stop and I agree that the economy is on tenuous footing at best. But until we get real interest rates at least solidly in positive territory, we'll never get the savings rate up closer where it needs to be to give a legit chance at a soft landing.
Posted by: RH at September 19, 2005 08:44 PM
The NYMEX crude prices have settled at a $60+ level even without storms. The oil prices have been volatile and will remain so. Natural gas is worse. There has been a sharp upward tendency here, too. This affects electric power prices, industry and housing and there is nothing much that could be done. The situation might lead even to physical shortages. Natural gas is depleting in North America and hurricanes have only aggravated the situation.
The consumer sentiment index as such doesn't mean much yet. It is normally a bit volatile and every downturn doesn't signal a recession. Hurricane effects will wane but those longer term trends will stay. The worries will be in this context.
Watch the natural gas, electricity and housing. There is definitively a housing bubble and they always burst. This, aggravated by rising commuting costs, could be the triggering event.
But would still guess that the Fed could handle the situation with the help of China. As long as China keeps growing at this pace it will drag the rest along somehow. Nevertheless the housing bubble is the weak spot.
Posted by: TI at September 20, 2005 05:47 AM
maybe in a machiavellian respose the frd passes today and the long end of the treasury market responds with a violent down trade and a significant steeepening of the yield curve as thr bond market vigilantes respond with dismay to the fed's ousillanimous action......but the law of unintended consequences kick-in and higher back end rates serve to slow the housing boom and give us a soft landing. jjj
Posted by: jjj at September 20, 2005 05:56 AM
i am working and didnt have time to proof read the previous posting....my apologies for a sloppy job.......i hope you get my point though.jjj
Posted by: jjj at September 20, 2005 06:09 AM
Housing bubbles doesn't have soft landings. The housing prices growth gan slow down but if the tide turns they will fall rapidly. Everybody wants to wait for still lower prices. FED has a good pretext to pass now. It will not be unexpected. FED can do a lot but it cannot control natural gas prices. So we will see.
Posted by: TI at September 20, 2005 07:54 AM
I'm curious as to what your reasoning is as to why housing prices must fall rapidly if at all. I'd have thought on balance they'd resist depreciation more than say equities because they're less tradeable.
So I'm curious to hear the counterargument.
Posted by: RH at September 20, 2005 08:27 AM
If there truly is a housing bubble, then speculators have driven prices up anticipating future gains. If appreciation doesn't fall but merely slows below the holding costs, then they should rationally dump their investments, precipitating falling prices. Speculators usually convince themselves they are investors though and wait until fear or necessity overtakes them which exaggerates the swings. Those who have not purchased for speculation and have the capability to hold on generally will which limits the magnitude of any decline.
Posted by: Lord at September 20, 2005 09:38 AM
Rapidly bursting housing and real estate bubbles are not rare. "Lord" has right, during a bubble, ie. when real estate prices are rising rapidly, investors get their profit from the increasing prices, not from rents. They dump if the prices stop rising.
But there is a more important effect. Households bying homes are willing to pay high prices if the price level is rising. Rising price gives them security because they think that they can always sell their house at a profit and pay their debts easily. They see the present price as a bargain compared to future prices and tend to make quick deals.
When price trend changes, sellers try first to get the "right", higher price and wait. This creates a glut. Buyers are in no hurry, soon they see that they can get discount and a better deal. Now the price is compared to future lower prices and every price seem too high. Debt becomes risky because selling the house will mean taking a loss. Real estate loses its value and mortgages go bad. Foreclosures and panic selling worsen the situation. Many have to sell but byers will wait for still lower prices. The bubble bursts, often violently.
The prices will settle in the end to a level corresponding average building costs (and long term land prices). There is a bubble when housing prices rise considerably over that level and the pricing and demand is driven by rising prices, as in a typically speculative bubble. If there really is a bubble it will inevitably burst. No soft landing. The above mechanism will ensure that. The bubble may be long lived but because the prices cannot rise indefinitely, it will burst as soon as something triggers the price trend change.
This is well known and have seen it happen (I do not live in the US).
The housing bubble is disastruous because people have used to believe that a home has a tangible value, it is a "real" investment, "real estate", and cannot lose its value. Housing bubbles start slowly and strengthen the belief that real estate prices go only up and there is no risk. That is why the bubble can live long and get big. There is all the time somebody warning that there is a bubble but these warnings are of course wrong and serve only to strengthen the belief that there is no risk. So the bubble grows. And when it bursts it will ruin many people. So it goes.
Posted by: TI at September 20, 2005 10:29 AM
Lord, where do you think we are in that process? I am a seller (Westchester, NY) and wouldn't you know, we are seeing a glut. Buyers are indeed showing patience. However, I don't see any evidence of price declines, while we are seeing some accepted offers lower than asking it is not too significant yet.
Obviously this is not a real estate blog but the question seems begging for data points.
Posted by: Paul at September 20, 2005 11:25 AM
There have been many real estate bubbles in crashes over time, with the 1837 panic being triggered by a crash in cotton land prices after a bubble, and the 1925 Florida land bubble being another biggie. More recently Boston and San Francisco and some other areas have seen sharp declines such as at the end of the 80s.
However, sometimes they are not such sharp declines, although maybe it would be better if they were. In the latter category is Japan, where prices have fallen at about a 3% rate per year since 1992, although with some evidence recently that at least in Tokyo they may have finally stabilized. That case is peculiar in having a slower decline than the runup was, whereas usually speculative bubbles crash harder than they go up. One reason real estate is more likely not to crash so hard is that people can hold them off the market and rent out if they cannot sell at a good price, although some may have to sell, especially if they are heavily indebted speculator/flippers.
Also, some bubbles can simply peter out in a cessation of increases with no real crash. For a useful discussion of historical bubbles, see Charles P. Kindelberger's _Manias, Panics, and Crashes_, now out in a fifth edition.
Posted by: Barkley Rosser at September 20, 2005 12:35 PM
The glut is a warning sign. The expectations are of course important. A long, seemingly robust real estate bubble makes people believe in rising prices and can save the bubble even when prices stagnate a bit. "Buy the dip".
Some kind of trigger is needed to burst the bubble. Rising energy prices and falling consumer sentiment may do the trick. Consumer sentiment may tell that belief in ever-rising housing prices is weakening. The prices will start falling if people don't expect them rising any more. They can wait. But this comes as a surprise. It is very difficult to predict.
Housing markets are different from equities because so many people are involved and many of are at the margin. For many taking a big mortgage is really a risk and they know it. Many have a near 100% mortgage so they may lose all if the prices go down. So panic selling is near. This adds to volatility.
Posted by: TI at September 20, 2005 12:48 PM
I'm sorry, I can't get HTML formatting in comments to work, so I'm going to have to refer readers to my post on the topic, which relates the confidence report to a focus group I attended:
Econbrowser, if you could advise to what I should be doing to create para and line breaks (traditional tags aren't working), I'd appreciate it. Your blog is very strong.
Posted by: Tom Blumer at September 20, 2005 01:59 PM
Oops, it works when posted but not in draft mode. Will remember in the future.
Posted by: Tom Blumer at September 20, 2005 02:09 PM
Prof. Hamilton has a number of posts about the supposed housing bubble situation at:
He is pretty skeptical about all these bubbles that people see all the time. After all with interest rates as low as they have been these past few years it is understandable that housing demand would increase. And when those rates rise, demand will fall and prices may fall with them. That won't mean we were in a bubble.
Posted by: Hal at September 20, 2005 08:38 PM
"He is pretty skeptical about all these bubbles that people see all the time. After all with interest rates as low as they have been these past few years it is understandable that housing demand would increase"
Bubbles really do exist and they do burst. This is a fact. But every price increase is not a bubble. That is also a fact. Housing bubbles are usually (if not always) connected to low or negative real interest rates and burst when the rates rise. But they have also clear speculative features that make them behave like speculative bubbles and cause the hard landing.
One sign is the affordability of housing. If the home prices are rising considerably relative to incomes the risks increase. People can afford housing they cannot afford because of the wealth effect of rising prices. I think we can find indications that there really is a bubble in the real estate. And if there is a bubble it will burst, no matter what.
Posted by: TI at September 21, 2005 12:06 AM
(remove 'at' to reply by email)
Remember with housing it is *real* not nominal interest rates that matter.
A fall in nominal interest rates simply pushes the burden of owning a home to the end of the mortgage life not the beginning (the principal is not so discounted away by inflation when it is repaid).
So it is low *real* interest rates that allow an upward structural shift in the price-to-avg household earnings rates.
If we believe that real interest rates are anomalously low because of Chinese Central Bank policy, then the setting for a bubble bust is clear.
As Andrew Smithers points out, real interest rates have averaged about 3% over a very long period (not sure how he is measuring in absence of a TIPS market pre mid 1990s). Right now they are closer to 1.7% in the US. Can that gap sustain?
Housing is also the classic case of a sticky market (high transactions costs, durable asset, sellers have an alternative which is to rent out or to continue to live there). Prices don't fall, but transaction volumes do, at least until the pain becomes unbearable (due to unemployment or rising interest rates). So you can have a very long slump which just looks like a market going nowhere.
Posted by: John at September 21, 2005 01:55 AM
John, you have absolutely right on the real interest rate. And housing markets are stickier than than the stock market because there they handle real objects. But this doesn't prevent steep price falls when the conditions are right. These things really happen and have happened before. In certain situations the stickiness contributes to the price fall. Those who have to sell are forced to accept any price, however low.
Now the question is, of course, whether there really is a housing bubble in the US or not. Housing bubbles are not absolute but relative phenomena. You cannot tell it from the absolute price level. And I'm not sure but suspect that there really is a bubble. Here we come to the present discussion in this blog. Rising energy prices, rising rates and falling consumer confidence can have the most tangible economic effect through the housing markets.
Posted by: TI at September 21, 2005 02:40 AM
I think that the definition of a 'bubble' has to have an element about *valuation* not just price.
So the tech bubble was signalled by an unprecedented rise in the PE of the stocks in question.
Similar valuation metrics for housing markets are harder to come by. On two that I know of:
the ratio of house prices to average household earnings
and the rental yield of housing (lower is more expensive)
the housing market in most major US markets (and most Anglo Saxon ones) looks very, very inflated relative to historic ratios.
To argue that that is not a bubble you have to argue that the world has changed from history. Which of course is precisely what dot com investors argued at the peak of the tech bubble. 'This time, it's different' are very dangerous words.
We don't actually know what caused the final deflation of the tech bubble (the Lastminute.com float and the Barron's article are usually identified as the culprits in the sense that they occurred just as the bubble peaked): the market just seemed to decide that actually, these stocks were overvalued, and so it started to sell. Positive stock price momentum became negative stock price momentum.
Posted by: John at September 21, 2005 03:15 AM
it is interesting to note that the current print of the CCI uses data gathered BEFORE katrina. Why did you not note this in your post? it seems relevant to me.
Posted by: sampo at September 23, 2005 04:07 PM
I'm not sure that you're correct about that, Sampo. It is true that the latest value of the index reported at http://www.sca.isr.umich.edu/ and at http://research.stlouisfed.org/fred2/categories/98 is the value for the August index of 89.1. However, my post and graph are based on the September index as carried by Reuters on Sept. 16, quoted by Macroblog as follows:
U.S. consumer confidence plummeted to a 13-year low in early September, battered by record gasoline prices and the full force of Hurricane Katrina, a report showed on Friday...
The University of Michigan's closely-watched consumer sentiment index eased to 76.9 in September from 89.1 in August.
Posted by: JDH at September 23, 2005 08:59 PM
Frankly, I welcome a decline in consumer "confidence" and even more so in expenditures.
The economy needs to shift to adding productive infrastructure and away from frivolities as a long-term trend since the capital requirements for new energy capacity will be large, having been deferred for so long.
As to housing, do this gedanken experiment. If the price of a middle class home in desirable California locations (SF, SD, Marin, Silicon Valley, San Clemente, etc) fell by a half overnight, would people rush in to buy? I'd say they would and not just on spec. That said, rental prices are being nudged downwards due to houses snapped up on spec and needing cash flow.
Hence, I'd wager that a good hit of unemployment would shake out some speculators and temper prices but it wouldn't collapse the overall market.
Posted by: Joseph Somsel at September 25, 2005 11:14 PM