January 04, 2009
Forecasted GDP in the New Year
The description of the consensus that growth will resume around mid-year -- while accurate -- does not convey much information about what is the consensus regarding the depth of the recession. Nor does it convey the degree of disagreement regarding the timing and strength of the recovery. To provide some isnight , here is the mean forecast for GDP into the new year, according to the WSJ's December survey.
Figure 1: Log real GDP, from 25 Nov 08 preliminary release (blue), potential GDP (black), WSJ mean forecast from December survey (red), high and low forecasts (teal), and third highest and third lowest forecasts (green). Source: BEA NIPA release [link], CBO estimates of 9 Sep 08 [xls], WSJ survey of forecasters from December [link].
The mean forecast indicates a recovery (i.e., resumption of positive growth) starting in 2009Q3. Despite the positive growth projected, the output gap implied for end-2009 is 6.4% in log terms. I've also depicted the range of uncertainty by showing the trajectories implied by the top (James Smith/W. Carolina U. and Parsec) and bottom (Dana Smith/Comerica), where the ranking is made on the basis of the average growth rate over the 2008Q4 through 2009Q4 quarters.
The range of estimates is quite wide, but is driven by outliers. Taking out the top two (the usual suspects, of Mark Nielson/MacroEcon Global Advisers; James F. Smith/Western Carolina University & Parsec Financial Management) and the bottom two (Dana Johnson/Comerica Bank; Douglas Duncan/Fannie Mae) leads to the trajectories indicated by two green lines (low: Maria Fiorini Ramirez & Joshua Shapiro/MFR, Inc.; high: Richard DeKaser/National City Corporation). This implies substantially more agreement about the prospects for the economy, although the implied output gap for this low estimate is still 8.2%.
Posted by Menzie Chinn at January 4, 2009 04:00 PMdigg this | reddit
Much too optimistic all around, Professor.
We have just completed year one of the five year downturn.
Real GNP fell an average of 10% over '30-'33, so that is my projection for '09 GDP.
We may be amateurs, but we are better than the pros: back in March '08, my wife, kids, and I made a friendly wager on year end values for gold, S&P 500, Nordstrom, and Costco. We were within 5% on three of the four (we were too optimistic on Nordstrom, which dropped much further than we expected it would drop).
House prices have already dropped by twice the amount that they dropped in the '29-'33 Lesser Depression. The risk premium, as measured by Baa - Aaa from the St. Louis FRB, has risen to levels last seen in Sep. '31.
We are in the second innning of the Greater Depression, Professor.
Posted by: jg at January 4, 2009 05:36 PM
How many of these prognosticators predicted the collapse in the first place, even so recently as Jan'08 when the recession was already a month old?
Posted by: bryce at January 4, 2009 08:14 PM
Considering that the outliers have ruled the day for the last two years, I believe it is incorrect to remove them. Most of these analysts are still on their business-as-usual la-la land, as are most folks on Wall Street and in Washington.
This just confirms to me that we have a long, long way to go.
Posted by: L'Emmerdeur at January 5, 2009 05:28 AM
There's the lack of confidence in the consumers...or izit partly their empty pockets? and then there's the lack of confidence in the students (that B us) in the (former but still active...possibly performing, possibly posturing) authorities (professors, professional economists, brave blog holders and more)...
and so...to jg's comment where he steps like no commenter has ever stepped before:
"Much too optimistic all around, Professor.
We have just completed year one of the five year downturn.
Real GNP fell an average of 10% over '30-'33, so that is my projection for '09 GDP"
But betting iznotit. We care not how lucky you bin before...no, we care about the calculation, how you, jg, arrived there...so that we might learn this...thinking...and how it might prepare us (and friends...and even enemies, now that I'm learnin the cheekiness of itall)...in the event that 2009 is different from 1931.
How would you like it if I were to diss yours with:
"Much too optimistic all around, jg.
We have just completed year one of a 15-20 year downturn following the more recent Japanese experience. "
Well, I'm sayin liking it should have the usual baggage of reasoning with it, behind it, backing it, The Support. No matter how unfashionable that might be at the moment, it is still preferable (for good reasons, not fashionable ones) to the shameless marketing...which got us here in the first place...lest we forget. Remember? REMEMBER?
Posted by: calmo at January 5, 2009 07:57 AM
If the goal is to figure out who has the right forecast, then pitching the outliers is the wrong answer. Oh, and good luck picking the best forecast.
If the goal is to convey the "consensus regarding the depth of the recession" as our host says is his intention, then to the extent that the forecasts are not normally distributed, pitching the outliers may help. More or less the same answer to jg. I'd also say that getting a few forecasts right one time is among the worst things you can do for your objectivity. Thinking you are better than some particular guy because you were closer to the right answer than a group of guys? Rethink that.
Posted by: kharris at January 5, 2009 08:03 AM
It'll be interesting to see what the December consumption figures show. November recorded a big increase, due to the collapse of commodities prices, retail sales since then point to a further increase. If PCEs don't fall as anticipated, consequent very rapid deflation increasing real disposable incomes, then the gross declines anticipated will be anything but.
Another point that I'd be interested to hear people's views on is how does the fall in prices impact on the PMIs? These show a fall in the value of orders at current prices, but if the fall in prices is large enough, then this will significantly offset the scale of that decline giving a false reading surely?
Posted by: bill j at January 5, 2009 08:12 AM
calmo and kh-: I not only prognosticate, I bet my prognostications.
I forecast that the S&P would fall to 900, and triple shorted the market. I put every single cent, plus a margin loan, on that bet. I closed out my shorts in early December (having grown my money from X to 3X), and will leave the workforce on Jan. 30, to concentrate on protecting our newfound nestegg and preparing for the social chaos that will accompany this Greater Depression.
Here's the logic on my '09 forecast (it's all about paying off/writing off household debt).
Last depression -- debt problems were with mortgages and margin loans
Boom: '22-'29, mortgages, consumer credit, and margin loans ('MCM debt')/GDP: 54% --> 69%
Bust: '29-'33; MCM debt fell 8% per year and GDP fell 12% per year --> ratio of MCM debt/GDP grows to 124% in four years
Recovery: '33-'39; MCM debt remained flat and GDP grew 10% per year --> it takes until '39 for MCM debt/GDP to fall to normal level of 51%
This depression -- debt problems are with mortgages and consumer credit
Boom: '80-'07, MCM debt/GDP: 48% --> 104%
Bust: '08-'12, assume that, again, MCM debt falls by 8% per year and that, again, GDP falls by 12% per year --> ratio of MCM debt/GDP grows to ~125% in five years
Recovery: '13-'22; assume that, again, MCM debt remains flat and that GDP grows by 10% --> it takes until 2022 for MCM debt/GDP to fall to normal level of 48%
15 years in total: five years of downward movement and ten years of recovery until we reach normal levels of MCM debt/GDP
I hope I am wrong. But, so far, I see nothing that shows that I am.
Posted by: jg at January 5, 2009 09:15 AM
The economic consensus on just about every upcoming economic datapoint from weekly unemployment numbers to annual GDP or market projections over the last year was wrong, estimating performance well above reality. I do not see this changing in 2009. Indeed, I will only begin to look for an economic turnaround when the consensus starts consistently overstating the negativity in the economy.
In the meantime, I will listen primarily to those who have been right so far, starting with Roubini.
Posted by: Terry at January 5, 2009 09:24 AM
jg - That's an interesting analysis but it's incomplete, in that the US citizenry is also burdened with corporate (~$25tr) and public sector debt ($~15tr). Both of these debt figures are substantially higher than in the late 1920s. While public policy has undoubtedly been improved since that time, the gross US fiscal situation is even more dire. So the situation may be even worse than in 1929.
Posted by: jult52 at January 5, 2009 10:15 AM
Just to add another note of scariness, the debt figures I just provided are approximate. As far as I can tell, no one has good information on what the debt load is, largely because of derivatives and exotic financial instruments. And no one knows how the debt is distributed geographically. Unless they are privy to unpublished information (which I admit is very possible), regulators are making stabs in the dark concerning the aggregate US money supply and debt situation.
Posted by: jult52 at January 5, 2009 10:22 AM
US debt is probably worse than in the 30's, even without considering looming unfunded liabilities such as Pendion Guaranty, FDIC, Medicaid, SS, et al.
On the other hand, we can hope that the new administration will not be quite so stupid as FDR's was with the likes of the NRA, the extreme taxation of retained Corp. earning, raising income taxes steeply, etc.
Posted by: bryce at January 5, 2009 11:34 AM
jult52, I agree with your points, that the overall debt -- incorporating corporate and government debt -- load is much higher than in '29. IMO, if households had normal debt/income levels, high corporate and government debt would not be a problem. Thus, I think the linchpin is household debt.
I see no ability for the Feds to repay the debt incurred to-date or the promises underlying all the bailouts and entitlements.
So, I foresee a reneging of the debt and effective collapse of the federal government. No way will ever-increasingly out of work Americans be credible, or effective, debt slaves for the Chinese, Japanese, and sheikhs.
I see a big blow-up happening in six to 18 months: tumbling tax receipts in the face of skyrocketing claims on government.
I hope that I am wrong.
Posted by: jg at January 5, 2009 12:21 PM
I have found that when dealing with estimates that averages are amazingly accurate. I went to a party where there was a jar of jelly beans and everyone at the party wrote their guess down in a list. When I made my guess I did not even look at the jar. I simply did a quick average of the guesses I could see on the list and wrote down that number. Yes, I won, missing by 3 jelly beans out of about 385.
But, I wondered, how accruate would be the average if people were allowed to eat the jelly beans, or to add jelly beans as they wished? And what if they were even allowed to change the size of the jar as they did this? This is much more closely related to our current economy than the fixed jar and fixed number of jelly beans.
I would say that the estimates of when the economy will turn are virtually meaningless. It is almost totally dependent on how much damage the Obama administration will do to the economy. If it is anywhere near as bad as the damage the Bush administration did, not one projection of a turn in 2009 will even be close.
Right now I am not getting a warm fuzzy from the Obama administration. They are talking tax cut but not supply side tax cut. Rather than a tax cut it will actually be a Keynesian spending program intended to stimulate not the supply of goods but the consumption of consumers. The result will be taking resources and capital from producers and redistributing it to consumers. Tthe result will be reduced production and distorted supply, as staple consumer goods are consumed and luxury goods sit unsold. Shallow Keynesians analysts will claim this is "classical" free market over-production and they will probably institute new policies that will only exacerbate the problem reducing the production and supply of all goods.
I hope I am wrong but I foresee a Keynesian crash coming as the Obama New Dealers try to micro-manage the economy.
jg, my compliments to the chef!
Posted by: DickF at January 5, 2009 12:34 PM
Before being too impressed by forecasts of any kind we should remember that all of six months ago -- July 2008 -- many very prominent forecasters were predicting oil at $150 to $200 by, well, today, and many economists were arguing the Fed should raise interest rates to pop the commodities bubble.
Remember all those arguments going on then about "speculators" driving up the price of oil?
What happened with that? ;-)
I'm not saying forecasts should be dismissed -- but when considering them keep a salt shaker at hand.
For all we know, all that Fed monetary stimulus could kick in a lot stronger than expected and a year from now the big concern will be whether the Fed can shrink its balance sheet and get the money supply restrained fast enough to avert a bad inflation. Would it be a bigger surprise than what's happened in the last year?
"It's tough to make predictions, especially about the future." The great Yogi.
Posted by: Jim Glass at January 5, 2009 01:53 PM
Thanks, DickF. We'll see if this fool and his money are soon parted: we are going all in in gold mining stocks and gold (all held in Switzerland), and will have some fun shorting the S&P 500 with a small sum.
I agree with your read of the incoming administration. I look forward to repatriating my capital and putting it to productive use, but only after I feel comfortable that there is a return to rational and honored rules, which schmuck GWB (whom I voted for twice) and his fellow shysters -- Paulson, Bernanke -- strongly pushed us away from this year.
Posted by: jg at January 5, 2009 03:26 PM
DickF: Take a statistics class and you will see how accurate/inaccurate averages can be.
Jim Glass: Predictions are tough to make. So many things in the economy are interrelated however, most models use a single outcome when modeled by economists. We have a dynamic system here that needs to be modeled as such. My main problem with peoples forecasts/predictions is that the lack of prediction bounds. Don't just give me an average, give me an upper and lower bound on what may happen. This lets me know how sure you are in your prediction.
I agree that the economy is very very fragile right now and I personally think that government intervention should be targeted. What we will probably get is throwing money at it and hope that it will work.
Tax decreases probably won't work. Why? Many many poor people who make up a large portion of consumers do not pay income taxes so you are not helping them. Middle class people will simply save the money since it is unlikely to be a large amount, ie enough for a downpayment on a car or even a new DTV.
Where the government can help is to start making banks lend money. If the fed all of the sudden started making mortgages and car loans for people and making interest on it, the banks would see that and want a piece of the action and would start lending too. Right now the banks seem to be in a odd sort of stand off. An analogy... Its like they suspect the food is poisoned and are waiting for someone to take a bite to see if they die. No one wants to go first. Let the federal government take the first bite and show them that there is money to be made and then greed will take over.
Posted by: ed at January 6, 2009 06:07 AM
Been there, done that. I have a lot of jelly beans. Used the same technique over Christmas and gave them to a friend.
Posted by: DickF at January 6, 2009 07:56 AM
The article was a report of the consensus of various econmists. Now we know what won't be happening. Let's see. the housing market hasn't bottomed with the Alt As and others to reset.I believe most people are correct when they think unemployment will get worse.
Double digits by years end is my guess. 2008 was the year that the financial meltdown (for instance I read in the Washington Post that AIG had stopped making unhedged CDS by 2006 but we all know the bill is up to 152 billion and hasn't stopped)started to affect the real economy. 2009 will be the year the real economy global depression will have knock on effects on the financials who will have further effects on the real economy... but the fact is there were too many car manufacturing plants in the world - more cars were being made than people could afford to buy. There was more steel capacity in the world... because FIRE had stopped being the place where surplus wealth was wisely distributed (was it ever?) and became a place where more and more money in tremendously complicated maneuvers was being skimmed off the top.
Kenneth Rogoff, former chief economist of the IMF 2001-2003:
The financial services industry had been taking in 30 percent of corporate profits and 10 percent of wages despite representing only 8 percent of GDP (at its peak, and that is counting insurance). Why should a supposedly efficient financial system be soaking up so much of GDP? It is quite possible that a lot of what has happened to our overbloated financial system needed to happen anyway, albeit one would have expected the process to take five years instead of five days.
Anyway, to me the real question is whether Helicopter Ben and the new Keynesians of Obama taking over from the failed Chicago schoolboys will be able to use a fiat money system to stave off the deflation now happening which has followed the inflation that was building globally last summer. The outside the box bloggers consensus it that more writedown are sure to come. Some think (goldbugs) the deflation will be followed by hyperinflation of the US dollar. Some think Helicopter Ben will be nimble and quick to know when enough is enough. How they can think that when - Federal Reserve Bank Chairman Ben Bernanke's Testimony to Congress 3-4-08 “Clearly, the U.S. economy is going through a very difficult period. Much necessary economic and financial adjustment has already taken place, and monetary and fiscal policies are in train that should support a return to growth in the second half of this year and next year.”
“I remain confident in our economy's long-term prospects,”
I think the best we can hope for is that the economic stimulus (if its done well by Obama and congress oh well) and the 'quantitive easing' of the FED will put off the bottom until 2010.
Posted by: Gepay at January 9, 2009 06:04 PM