Friday, January 30, 2009

My Q4 National Income Forecasts Evaluated

The BEA and I agree very closely on the components of nominal national income (they have not calculated corp profits yet). In fact, they see a bit more national income than I did.

It appears that the main disagreement between BEA and I is about the inflation/deflation rate. They see -0.3 and I see -3.3.

Remember my forecasts

I had three real GDP forecasts ranging from -2.2 to +.8 (annual growth rate from Q4). The "experts" consensus was -5.4.

According to market participants at intrade.com, there was only a 10 percent chance that the BEA would report something this morning that was better than -4.0.

This morning we learned that the BEA's "Advance" estimate was -3.8. It will be followed in about a month by its preliminary estimate, and then later its final estimate. I mentioned this morning that they have more work to do on deflators.

I had previously predicted that real GDP "
might" go up, although I still said that slightly down was the more likely. I was very precise about that: I said that it was 1/3 likely to go up and 2/3 likely to go down.

There are no rules in the blogosphere, but it is factually incorrect to quote me as predicting that real GDP WILL go up. "WILL" and "MIGHT" are different in plain English (not to mention that I also put it terms of probabilities).

$120 per person

The BEA says that real GDP fell $120 per person in Q4.

Even if we attribute all of that drop to the banking crisis, that means the crisis' damage was $120 per person.

Yet, in the same quarter, Congress dedicated $2300 per person to bailing out banks. Looks like we got ripped off.

I am suspicious about the Investment deflator

This morning the BEA said that price of investment goods went up in Q4. This was especially true for nonresidential structures (7.4% annual inflation rate) and for equipment (3.4% annual inflation rate).

Their equipment inflation estimate may hold up -- the PPI for capital equipment went up every month of Q4.

I am dubious about the structures inflation. The PPI for construction went down every month this quarter. I would think that residential and nonresidential construction prices would move together. Fixing that could turn today's estimate of -3.8% per year GDP growth (for the quarter) into a revised -3.5%.

Wow I was close on Nominal GDP

Wow, in dollar terms, I was close on each of the spending categories (see below)!

I expected the GDP deflator to fall a lot more, and the experts thought it would rise. I wonder how much GDP deflator revision is possible.


Flashback: Q4 GDP deflators

The BEA released their deflators today:

GDP deflator: -0.3 percent (I had predicted -3.3 percent)
PCE deflator: -5.7 percent (I had predicted -4.6 percent)
GPDI deflator: +0.5 percent (I had predicted -0.8 percent)

My nominal GDP prediction was close -- it's just that I saw the GDP deflator falling more. The experts saw the GDP deflator going up.

Productivity Surges in Q4

From 2008 Q3 to 2008Q4, aggregate labor hours fell 7.4% (at an annual rate).

The BEA reported today that aggregate real GDP (that is, aggregate spending) fell only 3.8%.

That means:

  1. Productivity grew 3.6%, and

  2. We have to continue to question the common "wisdom" that spending drops are causing the employment drops. I tend to think that causes are bigger than effects: its the employment drop that causes the spending drop.


Flashback: GDP will not fall below $11 trillion

I promised that real GDP would not fall below $11 trillion (year 2000 chained). That is where real GDP was at the peak of the housing boom. If we stay above that floor, real GDP is doing better than the 1981-82 recession.

As you can see below, this morning's release of 2008 Q4 does not bring the $11 trillion mark much closer.


Wednesday, January 28, 2009

Fed Takes More Steps to Reduce Employment

I saw a report today that the Federal Reserve is taking still more steps to modify mortgages, and thereby reduce homeowners' incentive to work.

2008’s New Economic Highs Deeply Disappoint


When measured in terms of terms of financial gyrations and national employment, 2008 was an absolutely terrible year. When measured in terms of production, the U.S. economy in 2008 was the best in its history.

This week the Bureau of Economic Analysis will release its first estimate of Gross Domestic Product (G.D.P.) for the fourth quarter of 2008. G.D.P. measures the total amount produced and spent in the nation during a particular time frame, such as a year or a quarter-year. Some say that G.D.P. “is the best barometer of the country’s economic fitness.”

The first column of the table below shows GDP for the four quarters of 2007. 2007 United States G.D.P. was about $3,500 billion dollars for the United States in each quarter, for an annual total of $13,807 billion dollars. To put that more personal terms, the last rows of the table divide by the total population: 2007 G.D.P. was $45,776 per person.


One reason that more dollars can be spent in one year as compared to the previous year is that a dollar is worth less over time, but that by itself does not help American living standards. In order to compare 2008’s results to 2007’s in a way that is informative about living standards, most economists (including those at the B.E.A.) refer to real G.D.P.: dollar G.D.P. adjusted the effect of inflation.

The second column of the table therefore reports real G.D.P. for each quarter of 2008, measured using 2007 prices. That is, production for 2008 is measured using the items actually produced in 2008, but values each item at its 2007 price.

In each of the first three quarters of 2008, more was produced in our economy than was produced in the same quarter a year before. For example, G.D.P. in the third quarter (the months of July, August, and September) of 2008 was $26 billion more than in the same quarter of 2007.

We do not yet know what the B.E.A. will measure for the final quarter of 2008. Pundits such as the economic teams at Goldman Sachs or the Bush White House say that the growth rate of real G.D.P. from the third to fourth quarters will be about -5 percent per year, which means that they expect real G.D.P. for the fourth quarter will be 1.25 percent lower than it was in the third quarter: about $3,465 billion.

My estimates (detailed here) suggest that the B.E.A. will report fourth quarter real G.D.P. to be $3,503 billion – or about the same as in the third quarter. Despite our disagreement about Q4, the pundits and I agree that the annual total for 2008 will be about $14,000 billion, or very close to $46,000 per person. With real production of $46,000 per person, the “recession year” of 2008 produced more goods and services than did 2007, or any other year in U.S. history.

I have not seen any explanation of how the U.S. economy can reach new production highs at the same time that jobs disappear and its financial system is in chaos. But – given that the disease has such strange symptoms – I have to wonder whether the usual remedies are still appropriate.

Monday, January 26, 2009

Analysts Still Down on 2008 and 2009

"Many analysts predict the economy will have contracted at a pace of 5.4 percent in the fourth quarter when the government releases that report on Friday. If they are correct, that would mark the worst performance since a 6.4 percent drop in the first quarter of 1982."


Recall that my calculations for 2008 Q4 say -2.2, +0.4, and +0.8 (depending on the method) -- as compared to the -5.4 cited above.

Saturday, January 24, 2009

Supply Curves in "Slumdog Millionaire"

It is politically incorrect to suggest the supply of unemployed persons slopes up -- that an increase in payments to unemployed people will increase their numbers.

[warning: the laws of supply and demand sometimes reveal themselves in graphic and disgusting ways. You may not want to read what's below -- or see the film. The film is nominated for Oscars, so I assume that most of you already know about it]

Mysteriously, it is political correct -- fantastically popular, in fact -- to suggest that the supply of deformed children slopes up. "...hundreds of young children have had their arms and legs chopped off; scores of others have been blinded. The gangs also pour acid on to the children’s bodies, leaving them with suppurating wounds. A happy ending for the stars of the film Slumdog Millionaire - but for real slum dwellers the future is bleak. Their suffering comes down to one thing: money." (quote from here, but see the film yourself)

On the Relevance of Wartime Multipliers

Someone asked about the relevance of wartime multipliers. Just to be clear:

  1. I have written a lot about crowding out, and little of my analysis relies on wartime observations.
  2. It used to be said that the wartime spending pulled the U.S. out of Depression. Given that President Obama is telling us that he'll pull us out of recession, it is of some relevance that the wartime spending still crowded out private spending. Thus my link to Professor Barro who, incidentally, "wrote the book" on how government spending affects GDP.
  3. Even if it were true that the U.S. economy was at full employment before WWII began -- in which case it might not be a valid comparison to 2009 -- I have already directly addressed (in at least two different ways) -- the myth that the stimulus plans in Washington would actually utilize currently unemployed resources.
  4. Conspicuously absent from the commentary of those who claim that public spending would stimulate private spending are actual examples in which that actually occurred. [Admittedly, those economists (are there any besides Professor Krugman?) are vastly outnumbered and therefore need more time to put their case together]

Friday, January 23, 2009

Look Out U.S. Senate -- Here Comes Emma Alumn

Ms. Gillibrand was my wife's high school classmate at Emma Willard (she was called Kirsten Rutnick back then).

She blanketed a sprawling Congressional district in upstate NY to unseat the entrenched incumbent.

She voted against the Wall Street bailout.


British economy

I have not studied the British economy, and have no idea whether the specifics of their economics are similar to ours.

I noticed that their Q4 real growth rate was -6% per year, after being -2.5% per year in Q3, 0% per year in Q2, and +1.4% per year in Q1 (that is, their cumulative growth through Q3 had been -0.4% per year).

Our real growth rate through Q3 was +1.0% per year. Given that our economy was doing a lot better going into Q4, it seems that we should also do better in Q4.

CPI 1929 and 2008: Professor Nunes

Professor Nunes makes a compelling point about 1929 deflation: he says it came from reductions in the supply of money, whereas 2008 deflation (he says) came from relative price changes.

I think that 2008 deflation also came from a reduction in the supply of money -- relative to DEMAND. The supply of money did not fall in 2008, but the demand dramatically increased (the "flight to safety") and the Fed failed to fully accommodate it. It's the Fed's job to keep money supply in line with money demand. But read his document and let us know what you think.

Thursday, January 22, 2009

CPI in 1929 and 2008

It is Bernanke's job to promote price stability. How is he doing? Let's look at 1929 and 2008.

I have normalized Oct 1929 to be 100 (the stock market crashed at the end of that month) and Oct 2008 to be 100 (the stock market crashed at the beginning of that month).


Defenders of Bernanke can blame "falling commodity prices." That was the excuse in 1929, too. [In both 1929 and 2008, one could revise the CPI so that it excluded all items with falling prices, prove that the revised index did not fall, and thereby assign the blame for falling prices to the excluded items.]

Robert Barro Looks at Wartime "Multipliers"

Of course, they are less than one: public spending crowds out private spending.

Wednesday, January 21, 2009

Stimulation isn't Everything



President Barack Obama expressed a vision for federal spending: that 2009 presents an opportunity for the federal government to invest in infrastructure, health care, and education sooner rather than later. Investments like these have intrinsic value — they are appealing because they are believed to serve the greater good — but not much stimulation potential.

When the government hires employees or makes purchases, their ultimate economic effect depends on what private sector activity is displaced.

In one scenario – called “crowding out” in economic jargon – some (but not necessarily all) of the new government employees are people who quit their private sector jobs in order to accept better positions offered by the public sector, and many (but not necessarily all) the new government purchases just take items that would have been purchased by a household or private business.

For example, an improvement of public schools may cause some students to withdraw from private schools and enroll in public ones and cause some school employees to relocate from private to public. The stimulus potential is minimal in this scenario, because much activity is shifting from private to public rather than being recreated anew. Gross domestic product may increase, but less than the public spending does – because public spending “crowds out,” or displaces, private spending.

Many people believe that moving students from private to public schools has intrinsic value. Or that building a public highway has intrinsic value even if it reduces private construction. Or that modernizing health care data has intrinsic value even if it reduces some other activity in technology industry. Thus, government purchases might be desirable even if they are not stimulating.

The second scenario is called the “multiplier.” In this view, government hiring and purchases largely acquire employees and resources that would have otherwise been idle. If this were correct, than the government’s spending in one area would not reduce spending in another. Instead, it would employ otherwise unemployed people, and get them spending again, thereby creating private spending on top of the government spending – hence the term “multiplier”. In the multiplier view, G.D.P. increases more than public spending.

Despite the recent increase in unemployment rates, I see little reason why the multiplier scenario is realistic. For example, President Obama’s economists have explained how about half of the jobs they plan to create (both directly and indirectly) are jobs for women. But the large majority of this recession’s employment reduction has been among men. Thus, the Obama spending plan is not designed to primarily draw on the pool of persons unemployed in this recession.

President Obama has a vision to spend more on health care, largely for its intrinsic value. Its stimulation value is minimal because unemployment is low in that sector; health sector employment has actually increased every single month during this recession.

Even the construction industry shows crowding out. The chart below shows two types of construction spending over time: residential and non-residential. The housing boom prior to 2006 clearly crowded out non-residential construction.



Perhaps there was not enough unemployment during the housing boom to realize the multiplier effect. But in the two-year period since the end of 2006 – the last twelve months of which have been an official recession – shows a similar pattern: Reduced spending in one area allows for more spending in another.

Significant publicity has been given to residential construction workers and equipment that became unemployed since 2006, but less recognized is the absorption of many (but by no means not all) of those resources by non-residential construction projects. Public spending on construction that might come from the President’s spending plan will re-employ so of those unemployed, but it will also draw others out of their current work activity. The latter is why crowding out will occur even in construction.

Government spending will reduce private spending virtually anywhere it may be targeted. The case for government spending should be made on its intrinsic, not stimulation, value.

Tuesday, January 20, 2009

The Consumption Spending Myth

FACT: Consumption spending is 2/3 of Total spending in the economy.

MYTH: If consumption spending falls, then it is likely that total spending falls.

Below is a graph of Personal Consumption Expenditures (PCE) and PCE + Net exports, for the months April 2008 - November 2008. Net exports is another form of spending in our economy (and thus included in the total): net spending by foreigners (that is, spending by foreigners in the U.S. minus spending by Americans abroad).




The PCE series repeats what the press has told us ad nauseum: consumption spending had been falling since June (longer in real terms).

"Net exports" is a small component of total spending, so it is tempting to ignore it. But the bottom series shows that (through Nov) the PCE decline has nevertheless been offset by increases in net exports. November PCE + net exports is higher than every month this year, except perhaps June (according to the BEA, June was a mere 0.04 percent higher than November).

PCE has fallen in large part because oil is cheaper. But oil production is not our major industry, which is why we continue to produce even while we do not continue to spend (as much).

Government spending (not shown) has increased throughout the year, so that investment is the only spending category that is unknown for Q4.

Monday, January 19, 2009

My Q4 GDP Forecast Details

The pundits (eg., Goldman Sachs or the White House) say that real GDP will fall about 5 percent (seasonally adjusted at annual rates) Q3-Q4.

I have used three different methods to make my own forecast, and each of them says that the pundits are wrong. Here is what I find for seasonally adjusted real GDP growth Q3-Q4 at annual percentage rates (and thus in the same units as the -5 cited above):
  • Productivity based: -2.2%
  • Income based: +0.4%
  • Spending based: +0.8%

The productivity forecast is the simplest: I take the -7.4 %/yr decline in aggregate work hours and assume that productivity rises via a movement up the labor demand curve, which yields -5.2%/yr. The pundits appear to stop here, assuming no productivity growth for a given amount of labor. But productivity has been increasing throughout this recession, so why should it stop now?! Thus I get -2.2%/yr.

The forecasts based on income and spending are shown in more detail below. In terms of the income account, the fairly strong aggregate payroll spending (as indicated by BEA and by Treasury payroll tax collections) are my reasons for optimism. In terms of the spending account, my optimism comes from net exports and my view that this recession's investment decline will be MILD.

I do not want to exaggerate the degree of precision any of us have -- the gap between my most and least optimistic forecast is only about $75 per person. The gap between Goldman and my least optimistic forecast is also about $75 per person. But there is no doubt that I see Q4 a lot differently than do the pundits.






Note: I have been saying for a while (e.g., here) that real GDP may well grow Q3 - Q4. All that is new today is that I have organized the details well enough to post here.

Quibbles with Cochrane

Neither Professor Cochrane nor I believe that public spending will raise private spending. Both of us hate to see the taxpayer get ripped off -- monumental ripoffs have occurred recently and more are on the horizon.

But I quibble with him about a few details:

Professor Cochrane writes "Every job created by stimulus spending is one job not created by the private spending it must displace."

I agree very much that crowding out is real. WWII is a nice example. For a 2008 example, look here or in my post this Wednesday.

But he is wrong to claim that crowding out has to be one for one. Again look at WWII -- yes GDP went up less than government spending, but it still went up. The adverse wealth effect of taxes is one mechanism by which it can happen. An intertemporal substitution effect is another.

Of course, the fact that public spending can increase GDP does not make it desirable! But let's get the positive analysis right.

Professor Cochrane writes "The only thing that matters is new debt, not who takes the hit for old bad debt"

It DOES matter who takes the hit for old debt: if the hit taken is a function of income, then debtors have little (in some cases, zero) incentive to work. I have noted this several times as relates to "mortgage modification" and tax debt forgiveness.

Also note that construction spending (including nonresidential) is higher in Q4 than it was in Q3. How does this happen if savings-investment intermediation is so terribly broken as Cochrane says?

Friday, January 16, 2009

Industrial Production 1929 and 2008


Here's an update through Dec 2008. Pessimistic interpretation: 2008 is 1/3 as bad as 1929. 1/3 of the Great Depression would not be good.

My Forecast for Q4 deflators

The Dec CPI came out today: down .7 percent (at monthly rates) from Nov. The Dec 2008 CPI is a bit lower than the Dec 2007 CPI: 2008 was a year of (a bit of) deflation.

This gives me enough data to make a confident forecast for the 2008 Q4 GDP deflators that will be released by BEA this month. The forecasts are Q3-Q4 percentage changes at annual rates.

GDP deflator: -3.3 percent
PCE deflator: -4.6 percent
GPDI deflator: -0.8 percent

Thursday, January 15, 2009

Flashback: Real Growth and Nominal Reduction

The PPI for Dec 2008 is lower than it was in Dec 2007. The CPI probably will probably finish 2008 like that too (tomorrow we'll know more). Thus, for the 12 months of 2008, prices fell overall even while real incomes increased. A rare phenomenon indeed!

[It is forecasted that 2008 Q4 real GDP will be lower than 2008 Q3. But nobody claims that real incomes in December 2008 were lower than real incomes in December 2007, because the economy was growing for the first half of 2008]

Deflation Continues, Bernanke is Shirking

Price stability (or a stable low inflation rate) is Bernanke's only job. Even if you think that Bernanke should dabble in other things (like popping bubbles or promoting full employment), you have to admit that price stability is his main responsibility.

The BLS announced today that producer prices decline for the fifth straight month (they fell almost 2 percent Nov-Dec alone). For Q4, that's a 5.5% decline (quarterly rates). The annual deflation rate Q3-Q4 is 20%. That's Great Depression style.

Remember that the various dollar figures we've heard (eg., that dollar retail sales dropped 2.7 percent Nov-Dec) need to be adjusted for price changes. I expect the CPI deflation to be a bit larger than PPI deflation, so the drop in retail sales volume is not as severe as the dollar drop. Indeed, the November to December change in retail sales VOLUME was likely less than one percent.

[if you are one of those who thinks that Bernanke is permitted to ignore certain price categories such as food and energy, notice how much the inflation rate for non-food-non-energy has dropped in the last six months. Bernanke needs to get to work.]

Wednesday, January 14, 2009

Data released today

Retail sales were released today. We still don't know whether the drop was in volume or in prices -- Dec price indices are coming out later this week.

I suspect that real quarterly retail sales are 5.5% lower in Q4 than in Q3. They were 2.5 percent lower in Q3 than in Q2.

But remember that retail sales are a lot more volatile than total consumption (the former are about 3% of the latter). I am working on my consumption forecast, but it will probably be around -1% (at a quarterly rate).

The other news today is that import prices fell more than export prices: good news for us, bad news for China and the middle east.

Are Recessions Still Manly?



The terrible bottom line of Friday’s job report from the Bureau of Labor Statistics (B.L.S.) did not surprise many people – there were many fewer jobs in December 2008 than in the prior month.

Much less noticed, however, was Friday’s breakdown by sex of employment declines prior to December. Is it possible that one legacy of this recession is that women become a majority of the workforce for the first time in American history?

Years ago, women were a small minority of the workforce (outside the home). During much of the twentieth century – especially the 1970s and 1980s – women increased their share. By 1990, the workforce was 47 percent female and 53 percent male, according to the B.L.S. Many view this as one of the most important and desirable social and economic transformations of our lifetimes.

Until this recession, women throughout the 1990s and 2000s remained less than 49 percent of the workforce. However, that percentage has now passed 49 percent and may cross the 50 percent threshold for the first time.

In November 2008, the female workforce shrank more in percentage terms than it ever has in any one month – and more than ever over any single year – since 1964, if not longer. Nevertheless, female workforce reductions so far in this recession are smaller than those for the male workforce – even when measured in percentage terms – as has been the case in previous recessions.

The table below displays calculations for three recessions – 1990-91, 2001, and today. According to the National Bureau of Economic Research, the recessions began in July 1990, March 2001, and December 2007, respectively. The first two ended in March 1991 and November 2001. Two notes: First, nobody knows when this recession will end, and second, the most recent data for which the B.L.S. reports employment by sex is November 2008.



During the first recession, male employment fell by 2.0 percent, while female employment hardly fell at all (less than 0.05 percent). In the other two recessions, the percentage employment loss for men exceeded that for women by a factor of 3 to 5 (interestingly, although women still have a small minority of each recession’s employment decline, the female share of the decline has increased from one recession to the next, indicating that women’s jobs have become more sensitive to changes in the business cycle).

The bottom of the table shows how, as a result of the larger male jobs losses, recessions have increased women’s percentage of the work force by about 0.5 in recent recessions. For example, the 0.5 percentage point increase that occurred in the eight months of the 1990-91 recession is more than occurred for the ten years that followed, or in the six years that followed the 2001 recession.

If the pundits are right that this recession will be long and severe, then women may gain the 0.9 percentage points from November 2008 that would push them past the 50 percent milestone. Important milestones will remain to be achieved, but women’s surpassing 50 percent of employment is something that historians will note for years to come.

Real Growth and Nominal Reduction?

The BEA has quarterly national accounts back to 1947. During that time (62 years), nominal GDP NEVER fell while real GDP grew, even for a single quarter.

The BEA has annual national accounts back to 1929. During that time, nominal GDP NEVER fell while real GDP grew. Judging from the annual rates, it's possible that it happened for a quarter during 1933 or 1938.

[The BEA has monthly consumption back to 1959, during that time (599 months) nominal consumption fell while real consumption grew only two or three times: Nov 2008, Sep 2006, and maybe Dec 2001 (real consumption was essentially constant here)].

But it may have happened in 2008 Q4! Even I am confident that nominal GDP fell in Q4. But I have noted the significant (but still less than one half) probability real GDP grows in Q4.

Milton Friedman mentioned that it happened several times in history prior to the Great Depression -- does anyone have that reference? I suspect it is in his AEA presidential address, but maybe in the Monetary History.

[update (via Tino): p. 97 of the Monetary History says "From 1891 to 1896 net national product in current prices fell at an average rate of roughly 1 per cent per year, and in constant prices rose at a rate of roughly 2 per cent per year." Chart 16 seems to show some real income growth 1927-27, but a money income decline.]

Tuesday, January 13, 2009

Payroll Tax Revenues Keep Rolling in Q4

One indicator of labor market activity is the collection of payroll tax revenue by the U.S. Treasury, released today through the end of calendar year 2008. This has an advantage over other measures because it is comprehensive -- it counts all the dollars rather than just gathering a sample of employers. It is also a high stakes measure -- an employer has no reason to tell the Treasury that he is paying employees -- and then pay tax for them -- when in fact he is not.

A disadvantage is that Treasury collections are not seasonally adjusted, and can fluctuate due to quirks in the tax rules. For that reason, I focus on payroll tax collections for "Employment and General Retirement" because the rules for those are pretty simple as compared to the rules for individual income tax or corporate income tax.

Payroll tax revenue in Q4 (Oct, Nov, Dec) was $213.67 billion, as compared to $202.024 billion in Q3. That's an increase of 5.6% (at a quarterly rate)!

Note that, in the couple of years prior to 2008, payroll tax revenue fell a bit Q3-Q4 (in part, I suspect, because some workers go over the payroll tax gap in Q4). Thus, I suspect that seasonally adjusted payroll tax revenue growth Q3-Q4 would be even higher than 5.6%.

Another Reason Why Q4 GDP growth might not be so bad

Recall that GDP is the sum of spending on private consumption, private investment, government consumption and investment, and net exports.

The November trade report came out today, and suggests that Q4 net exports will improve about $50 billion (at quarterly rates).

How big is $50 billion? Well, GDP for 2008 Q3 was about $3600 trillion. Forecasters have been saying that GDP will fall at a 5 percent annual rate, which means that it will fall 5/4 of a percent from Q3 to Q4. 0.05*3600/4 = $45 billion.

Yes, you read that correctly -- the change in net exports (in the direction of raising GDP) is as large as the declines that the forecasters think they see.

We need to be careful because, in principle, an increase in net exports could just reflect a reduction in consumption (and thereby no change in the sum of consumption and net exports). This quarter, much of the trade deficit improvement is reduced oil imports. However, that is not a reduction in real consumption but rather a reflection of improved terms of trade (oil prices have dropped a lot more than the prices of goods we export).

I will soon finish my spending-based and income-based real GDP growth forecasts. I explained earlier that my employment based forecast is -2.5% (at annual rates). I expect the spending-based forecast to be at least as optimistic, and the income-based forecast to be the most optimistic of the three.

Monday, January 12, 2009

More Exaggeration of Great Depression Parallels

Obama's choice for CEA chair draws parallels with the Great Depression. She says: the economy got out of hand after the 1929 financial panic, and we should not let the same happen now.

But is today's economy -- so far without any "help" from the Obama administration -- following the same path? Below is the Federal Reserve's monthly Industrial Production Index (which also existed in 1929). I have normalized Oct 1929 to be 100 (the stock market crashed at the end of that month) and Oct 2008 to be 100 (the stock market crashed at the beginning of that month). One thing we notice is that, after the summer 1929, it was a long time until the index increased again (the first increase is off the scale -- in early 1931). In 2008, the index did not increase again until ... um ... October.


As in 1929, the industrial production index fell 4 or 5 percent PRIOR to the 2008 crash. In 1929, the index fell 5 percent in November after falling 2 percent in October. In 2008, the index fell 0.7 percent in November after rising 1.4 percent in October.

2008 looks different to me. On Friday the Fed will release December 2008 Industrial Production, and we can see how close it gets to the 1929 time series.

Nonexistent WMD V: Industrial production and Capacity Utilization

After Lehman failed and “credit markets froze” in the second half of September 2008, many people proclaimed that a second Great Depression would unfold. A few days ago, I explained how payrolls did not collapse as some had been predicting in late September and early October.

During the first days of October 2008, it was claimed that the frozen credit markets would especially paralyze businesses. Now that a couple of months have passed, let’s look at industrial production and capacity utilization (two measures of business activity that are available monthly) as measured by the Federal Reserve. The charts below show industrial production (measured as a index) and capacity utilization (measured as a percentage of full capacity) for each of the months of second half of 2008 (December data not yet available). Both fell in and again in September, and have remained higher than September in the months since.




The Lehman Brothers investment bank failed on September 15. The bank was deeply intertwined with other financial institutions – its failure to pay its obligations put its creditors at risk – and brought the credit crisis to its crescendo. By the end of the month, the intense financial chaos motivated Mr. Obama and others to warn that banks needed lots of money from taxpayers, or else businesses and consumers would get hammered. Because the bailout bill took time to pass, and then additional time for the Treasury to design and execute its $700 billion Capital Purchase Program (CPP), no bank received any money from the Treasury pursuant to the CPP until the last couple of days of October. Thus, if the warnings were right, business production should have suffered dramatically in October.

Industrial production and capacity utilization may put too much emphasis on manufacturing and thereby give a noisy picture of the overall economy, but we were promised disastrous results -- and thus results that would be obvious even in noisy data.

Politicians from both parties alarmed the American public at the end of September and in early October in order to justify spending $700 billion of taxpayer funds on a bailout of United States banks. I previously showed three measures of October and November activity that were by no means disastrous by comparison to previous months. Above are two more.

Sunday, January 11, 2009

Economic Numbers Coming out this Week

In the order of importance in my analysis:

  1. Dec real earnings (BLS, Fri)
  2. Dec Advance Retail Sales (Census Bureau, Wed)
  3. Dec PPI (BLS, Thurs)
  4. Dec CPI (BLS, Fri)
  5. Industrial Production (Fed, Fri)
  6. Nov International Trade (BEA, Tues)
  7. JOLT (BLS, Tues)

The Federal Reserve Bank of NY has a nice calendar.

Nonexistent WMD IV: Others are Calling it the "Great Hoax of 2008"

I wish I had thought of that!

These guys (here and here) have updated the banking data that I and Chari and Christiano and Kehoe had looked at back in October. The update fails to show an obvious footprint of a credit crunch.

Nonexistent WMD III: Small Business Income in October

After Lehman failed and “credit markets froze” in the second half of September 2008, many people proclaimed that a second Great Depression would unfold. A few days ago, I explained how payrolls did not collapse as some had been predicting in late September and early October.

During the first days of October 2008, it was claimed that the frozen credit markets would especially paralyze small businesses. Now that a couple of months have passed, let’s look at proprietor's income (a measure of small business activity) as measured by the Bureau of Economic Analysis. The chart below shows proprietor's income (measured in billions of dollars) for each of the months of second half of 2008 (December data not yet available). Aggregate proprietor's income fell in August. It fell about 0.5 percent in September, and has remained within 0.2 percent of that ever since.



The Lehman Brothers investment bank failed on September 15. The bank was deeply intertwined with other financial institutions – its failure to pay its obligations put its creditors at risk – and brought the credit crisis to its crescendo. By the end of the month, the intense financial chaos motivated Mr. Obama and others to warn that banks needed lots of money from taxpayers, or else small businesses would get hammered. Because the bailout bill took time to pass, and then additional time for the Treasury to design and execute its $700 billion Capital Purchase Program (CPP), no bank received any money from the Treasury pursuant to the CPP until the last couple of days of October. Thus, if the warnings were right, small business should have suffered dramatically in October.

Proprietor's income is a notoriously noisy measure, but we were promised a large collapse -- and thus one that would be obvious even in noisy data.

None of the above denies or confirms that our economy is headed for economic depression, because there are multiple pathways for getting there. Nor does it deny that the Treasury CPP might help in some way. But it does refute one of the scariest pathways to Depression – an immediate collapse of small business – that politicians from both parties alarmingly described to the American public in order to justify spending $700 billion of taxpayer funds on a bailout of United States banks.

Saturday, January 10, 2009

Obama team will not fix tough times for men

Male employment has fallen 1.8 million so far during this recession. If Obama is right that the economy will get worse before it gets better, we should expect that, sometime in the next few months, male employment will be down a total of 2.5 million (men lost 350,000 jobs in November 2008 alone).

Female employment is down much less (see my post next Wednesday).

The Obama team today released parts of a plan to save the economy and boost employment (they claim) 3.7 million by 2010 Q4 (almost 2 years from now), of which 2.1 million will be male employment.

The 3.7 million grossly exaggerates what government can do, but I am going to take their calculations literally for the moment. They are saying that male employment will still be down about 400,000 two years from now, whereas female employment will be a million higher than it was when the recession started.

Assuming that sex changes are not part of the fiscal stimulus plan (of course, they are not!), the Obama team admits that it intends to put to work a very different group of people than were working a year ago.

One defense of that part of the Obama plan is that the income matters -- it does not matter who earns it. I tend to subscribe to this view, but it implies that we should be looking at GDP and not employment.

If employment matters per se -- that is, public policy should "fix" low employment even if GDP were doing fine, then I do not see how the Obama team can defend putting people to work who were not working a year ago, and allowing those who worked a year ago to remain without jobs.

[added: a standard measure of employment is "payroll employment" -- that is, the number of positions on businesses' payrolls. Each position is either held by a man or a woman, so the "male employment" I described above is just the number of positions held by men. It does NOT refer to the number of occupations that are "male dominated", but rather to positions actually held by men. For example, some of the male employment is in nursing because some men are paid to work as nurses. Some of the female employment is in construction, because some women are paid to do construction work.]

January 30 will be a political and economic milestone

The BEA releases its advance estimate of 2008 Q4 real GDP on January 30. There is a chance (admittedly, less than 50%) that it will show that real GDP was HIGHER in Q4 than in Q3.

If so, the release will unleash a huge political and economic debate. Washington will be in the middle of the preparation of a massive stimulus package. The employment numbers (December employment is released Jan 9) will undoubtedly help the case for stimulus, but real GDP growth would make a great case against. How can we justifying a huge spending bill if GDP were seen to grow without it?

Friday, January 9, 2009

Review of My Forecasts

I have a number of forecasts. Most of them have already been shown to be true. The remainder need more time until the data needed to verify them are released.

The purpose of this post is to reiterate the forecasts and their economic or statistical basis, prior to the jobs report that is coming out in a few hours.
  1. real GDP will exceed 6 percent of its value at the time of the housing crash, sometime during 2009 or 2010. By comparison, private consumption will not increase as much.
  2. between now and then, real GDP will not fall below $11 trillion (year 2000 chained)
  3. between now and then, nonfarm payroll employment will not fall below 134 million.>

The basis for these 3: investment can continue even while banks are in a mess (see my nytimes piece in Oct).


(1) is not a short term prediction, because this basis does not say exactly how long it takes other sectors to absorb former construction and bank employees. (2) and (3) are both short and long term predictions.

If one of these turns out to be wrong, but investment remains strong then we will all be puzzled because it was the banks who were supposed to bring disaster through an investment collapse. But in this case I will admit that I neglected some very fundamental factors.

(4) housing construction will resume my summer 2009.

(4) is based on the combination of my view of the banking system (that it is able to make housing loans as soon as it makes economic sense to build houses) AND my estimate of the pre-"bubble" trend for housing demand. If I were wrong about the latter, then housing construction would resume later for reasons having nothing to do with banks. One way to confirm this interpretation versus others is to look at housing rents -- if they get high then I'll admit that banks are holding back supply.

(5) real GDP may grow Q3-Q4 less than -5 percent (annualized), but it is just as likely to grow more than zero.

(5) is NOT based on my hypotheses about the banking system. It is based on two things: (a) my view that productivity and employment move in opposite directions in this recession (this view has little to do with my view of banking and investment), which makes me a little more optimistic Q3-Q4 than the consensus forecast, and (b) my statistical analysis of October and November (all that I have right now) national income items. Thus, if (5) is wrong, it is because my productivity forecast was wrong or because the December data was dramatically different than Oct-Nov or because the national income accounts' "statistical discrepancy" (an official item) were large relative to the gap between my forecast and others'. The result of (5), and the reasons for it, will be known Jan 30, 2009.

Thursday, January 8, 2009

Federal Government Taking Steps on Mortgage Modification

Obama's stimulus package will likely allow "bankruptcy judges alter home loans in an effort to prevent foreclosures and urged other lenders to follow suit." (read more here)

This will likely encourage banks to follow the Streamlined Modification Program before a mortgage gets on a judge's desk. But I don't know the details or how judges would modify -- could there be some way this move reduces the now massive implicit tax from mortgage modification?

Wednesday, January 7, 2009

Flashback: Another Example of How Washington Helped Decapitalize the Banking Industry

A billionaire venture capitalist explains how he has not yet bought a distressed bank because the Treasury is doing all the buying these days.

Thanks Alberto for the tip.

October Hysteria in Hindsight


After Lehman failed and “credit markets froze” in the second half of September 2008, many people proclaimed that a second Great Depression would unfold. In hindsight, we readily see that at least one of the purported pathways to depression was never followed.

During the first days of October 2008, it was claimed that businesses would not be able to borrow from banks even for basic operational expenses, such as making their payrolls. As employees had to work without pay (so the story goes), the businesses patronized by those employees would suffer, and the downward spiral would continue. Then presidential-candidate Barack Obama said that “the credit market is seized up and businesses, for instance, can't get loans to meet payroll.” It was even suggested that “recession proof” employers such as colleges and municipalities would not be able to pay their employees.

Now that a couple of months have passed, let’s look at payroll spending measured by the Bureau of Economic Analysis. The chart below shows payroll spending (measured in billions of dollars) for each of the months of second half of 2008 (December data not yet available), including contributions to pension and health funds for employees. Aggregate payroll spending was highest in August, and has remained within 0.1 percent of the August high ever since.



The Lehman Brothers investment bank failed on September 15. The bank was deeply intertwined with other financial institutions – its failure to pay its obligations put its creditors at risk – and brought the credit crisis to its crescendo. By the end of the month, the intense financial chaos motivated Mr. Obama and others to warn that banks needed lots of money from taxpayers, or else the banks’ business customers would not be able to pay their employees.

Because the bailout bill took time to pass, and then additional time for the Treasury to design and execute its $700 billion Capital Purchase Program (CPP), no bank received any money from the Treasury pursuant to the CPP until the last couple of days of October. Thus, if the warnings were right, payroll spending should have been precipitously lower in October than in the previous months. Instead, payroll spending was almost $1 billion dollars higher in October than in September – not exactly the collapse that we feared.

Some of the details of the CPP became known earlier in October. Anticipation of the CPP expenditures cannot explain why payroll spending did not collapse in October, because the purported pathway to depression was about the day-to-day cash needs of otherwise strong businesses. Even the United States Treasury admitted last week that “capital [from its CPP] needs to get into the system before it can have the desired effect.”

Even when some (but by no means all) of the CPP funds finally got “into the system” in the last days of October and the full month of November, Congress was dismayed to see that banks were not using the funds to lend. Nevertheless, payroll spending did not collapse in November, either.

It is true that payroll employment fell by about 850,000 in October and November – and that’s serious as compared to the last couple of recessions – but the payroll spending data show that the employment loss was small compared to the spending collapse forecasted in early October. The fact is that more than 136,000,000 workers received their paychecks – more than $1.3 trillion worth in October and November combined – essentially the same aggregate payroll that was paid out in the two months prior to Obama’s warning.

None of the above denies or confirms that our economy is headed for economic depression, because there are multiple pathways for getting there. Nor does it deny that the Treasury CPP might help in some way. But it does refute one of the scariest pathways to Depression – a collapse of payroll spending – that politicians from both parties alarmingly described to the American public in order to justify spending $700 billion of taxpayer funds on a bailout of United States banks.

The Marginal Tax Rate Hike of 2009

I interrupt WMD week to announce the marginal tax rate hike of 2009:

AP reports "As the nation sinks deeper into recession, the IRS is offering to waive late penalties, negotiate new payment plans and postpone asset seizures for delinquent taxpayers who are financially strapped, but make a good-faith effort to settle their tax debts. ...those seeking help will have to demonstrate their inability to pay." [see also this report]

I explained in a number of previous posts (here, here, here, here, here, here, here, here, and here) how when mortgage debt collectors offer forgiveness, but only for persons with incomes that are low in comparison with their debts, that creates financial disincentives for working.

You might guess that the economics is much the same for tax debt collectors: when they offer forgiveness, but only for persons with incomes that are low in comparison with their debts, that creates financial disincentives for working. That's why I call that new IRS policy a marginal tax rate hike.

Tuesday, January 6, 2009

Labor Supply and Demand: Today and in the Great Depression

Figure 15 (from my new paper "What Caused the 2008 Recession? Hints from Labor Productivity") is a scatter plot contrasting this recession and previous ones along these dimensions. Each recession is one data point in the chart. The horizontal axis measures the change in the log productivity residual from one quarter prior to the NBER peak to the fourth quarter following the NBER peak. The vertical axis measures the change in the unmeasured labor distortion (also in log points). The first three recessions each had productivity shifts that were less than experienced during non-recession years. The 2008 recession is unusual in that it has normal productivity shifts (so far) throughout the recession, but the largest adverse labor distortion shock. The 1970s and 1980s had less increase in the labor distortion than did the other recessions.


The Great Depression of the 1930s (not shown in the Figure) is qualitatively like the 1970s recession, but multiplied many times. Cole and Ohanian (1999, Table 6) find total factor productivity to fall five percent in the first year of the Great Depression, and a total of 14 percent through four years. Mulligan (2005, Figure 4) finds the Great Depression labor supply distortion to increase 0.17 log points in the first year and 0.46 log points through four years.

In other words, 1929-30 would be in the same quadrant of Figure 15 as the 1970s recession, and on about the same ray from the origin, but five times further away. 1929-33 would be about 15 times further.

Monday, January 5, 2009

Construction spending so far higher in Q4 than in Q3

  1. November construction spending averaged $89.9 billion.
  2. October construction spending averaged $90.4 billion.
  3. Q3 monthly construction spending averaged $89.9 billion.
All of these are seasonally adjusted.

It seems to me that Q4 may well finish with more NOMINAL construction spending than Q3. Construction is also cheaper in Q4 than in Q3 --> more REAL construction in Q4 than in Q3.

Can someone explain to me where I can see this fall's credit crisis depress real aggregate activity? I don't see it in real payrolls (more on that Wednesday), real proprietor's income (more on that Thursday), or real construction spending.

The whole justification of the $700 billion bailout reminds me of the supposed WMD in Iraq -- scary but nonexistent. A payoll collapse, small business collapse, and construction collapse were somehow all prevented without the $700 billion.

WMD week

This week is my blog's WMD week. More accurately, it is politicians-said-there-are-WMDs-but-in-reality-there-are-none week.

President Bush said there were WMDs in Iraq, but nobody was able to find them.

President Bush, Presidential-candidate Obama, and several other Democratic party Senators, said on October 1 that an economic collapse was imminent unless the U.S. Treasury took $700 billion of our tax dollars and gave it to U.S. banks. The Bush administration and Congressional Democrats appealed to the same fears to rationalize giving several billion to the auto industry.

This week I will scour the economic data for October and November and try desperately to find those economic-WMDs. You will see that I won't even find a squirt gun!

Another Reason why Q4 GDP Just Might Increase

I admit that positive real GDP growth is an underdog. But here's another case for it:
  1. Prices have fallen 1-2% since Q3. Thus, nominal GDP has to fall at least 1% if real GDP is going to fall.
  2. Nominal compensation of employees and proprietor's income (at annual rates) is at least as high in Oct & Nov as in Q3. (The well publicized employment reduction should work through this item).
  3. Depreciation must be at least as high in Q4 as in Q3, because investment hasn't stopped.
  4. All of the above means that capital income has to fall at least $144 billion (at annual rates) Q3 to Q4 in order to make real GDP shrink Q3-Q4.
  5. $144 billion capital income reduction would far exceed the cumulative reduction in capital income since the beginning of this recession.
  6. $144 billion capital income reduction would increase labor's share to 0.683 from 0.659 a year earlier -- matching the largest YOY increase in labor's share since WWII.

Flashback: Nonresidential Investment Boom

I have consistently explained (eg., here and here) how the housing boom was diverting resources away from nonresidential investment. Once the boom was over, it became time to invest in the nonresidential sector again.

I was told that the latter couldn't happen, because banks were (supposedly) not lending.

Well, November showed yet another INCREASE in nonresidential construction spending:

"The Commerce Department reported Monday that construction spending dropped by 0.6 percent in November, less than half of the 1.3 percent decline economists expected. A 4.2 percent fall in housing construction was partially offset by a surprisingly strong 0.7 percent rise in nonresidential activity."

This was no surprise to readers of this blog!!

Construction spending is measured in dollars -- so it is an even bigger real increase (recall that investment goods prices have been falling these days).

Flashback: The Rhetorical Gap far Exceeds the Policy Gap

Before the election I blogged a couple of times (here and here) about how Democrats and Republicans talk so differently about public policy, but ultimately execute so similarly.

Since then, I noted how quickly President-elect and Democrat Obama withdrew is plans to repeal the Bush tax cut.

Now we learn that:
  • President-elect Obama plans to cut taxes MORE than Bush did!
  • President-elect Obama's Mideast stance may not be so different than we were led to believe. "Arab commentators and editorialists say there is growing disappointment at Obama's detachment - and that his failure to distance himself from George Bush's strongly pro-Israeli stance is encouraging the belief that he either shares Bush's bias or simply does not care."

You might say events have changed significantly since Nov 4, and these events coincidentally caused President-elect Obama to act more Republican. What events, exactly? Are we more worried about the economy now than we were in October (remember that's the month when we decided to give away $700 billion to banks)? Is it brand new that Israel has violent encounters with Hamas?

And why does coincidence perennially push in the directions of causing Democrats to act more Republican and Republicans to act more Democratic?

Saturday, January 3, 2009

The Economic Content of Politically Correct Business Cycle Analysis

  1. An economic tautology says that all market fluctuations are explained by some combination of preferences, technologies, and market distortions.
  2. A recession is defined (by NBER) to be a reduction in the quantity of employment.
  3. It is political incorrect to blame a significant part of recession on a reduced preference for working.
From these three, basic logic implies that either (a) the preference for working never falls, or (b) that it falls only at times when it is fully compensated by demand increases (that it, it only falls during non-recessions).

(b) seems like a remarkable coincidence, so we must conclude that the preference for working never falls.

Moreover, given that the preference for working never falls, the same logic implies that the preference for working never increases, or else (after 1000s of years of the human race) all we would do is work.

Political correctness makes labor market analysis easy -- we only have to study demand and distortions; supply can (and must) be ignored!

I really should stop fighting the utter neglect of supply, and instead appreciate its productive value.

Friday, January 2, 2009

Housing Price Changes this Fall -- Maybe Bad News for Banks

Case and Shiller released their October housing price index on Dec 30. It is a 3-month moving average (that is, the October value uses housing prices from Aug, Sept, Oct).

The Oct value of the index is 2.0% below the Sept value, which was itself 2.0% below the Aug value. As I predicted a couple of months ago, housing prices continue to fall. They will fall (and housing construction will continue to be essentially nil) until they are near construction costs. This should happen this summer.

Obviously, housing prices and construction costs can come into line in two ways: housing prices fall further, or construction costs rise. Construction costs stopped rising in October -- the October PPI for residential construction was actually 1.2% lower than for September. November's was another 1.3% lower than that. Unless we get some significant inflation in the very near future, it looks like owners of mortgages are going to see their loans decline further in value because falling construction costs are yet another force pulling down the value of their collateral.

A silver lining for banks is perhaps that markets already had anticipated some construction cost declines.

disclusure: long XLF

Thursday, January 1, 2009

Fundamental Inconsistencies in the Treasury's Report to Congress

Yesterday the U.S. Treasury reported to Congress about its $700 billion bailout program, which it now calls the "Capital Purchase Program."

It claims that the program helped "prevent a financial collapse" (p. 4). At the same time, it admits that “capital [from its CPP] needs to get into the system before it can have the desired effect.” (p. 5)

I find these claims to be fundamentally inconsistent, because (a) most of the CPP money has not yet been received by the banking system and (b) NONE of it had been received as of October 27. How did we survive between September 15 (when Lehman failed and brought on the crescendo of the credit crisis) and October 27? How did businesses make payrolls?

Maybe some credit goes to the Federal Reserve, who (from a different authority) was spending taxpayer dollars throughout September and October. But the whole justification of the $700 billion TREASURY authority reminds me of the supposed WMD in Iraq -- scary but nonexistent. A financial collapse was somehow prevented without the $700 billion.