Using truthiness to avoid a recession

"This economy is going to come on. I'm confident it will."
- President Bush

More than 3/4 of Americans think we are in a recession. Fortunately the Bush Administration has been able to prove the vast majority of Americans wrong this week, because all the numbers it released say that the economy is still growing.

Well, it is still growing as long as you don't look at the numbers too closely.

The headline number yesterday was Strong Consumer Spending. It was twice as high as market estimates.
However, there was a small problem with the numbers - they weren't adjusted for inflation.

Don't be fooled by a larger-than-expected increase in consumer spending. People aren't buying more - they're just paying more for what they buy.

Of course recessions aren't measured by consumer spending. They are measured by GDP growth (or lack of). The economy avoided a contraction in the 1st Quarter and grew by 0.6%, as was reported on Wednesday. Republicans were quick to seize on the numbers as proof of something or other.

So the headline numbers say that we are in slow growth, but not a recession. But below the headlines there are a few problems.

A buildup in business inventories, which bolsters growth in the period in which it occurs, helped the economy keep growing in the first quarter. Stocks of unsold goods rose at a $1.8-billion annual rate in the first quarter after shrinking at an $18.3-billion rate in the final quarter of last year.

Let me translate for you: the fact that companies kept producing goods that they couldn't sell was the reason that the economy growth in the last quarter. Now that's something to build an economy on, huh? And remember that the 0.6% is annualized, thus making economic growth in the first quarter of 0.15% (note: the population of the country increased nearly 1% in the first quarter).
Of course that wasn't the BIG reason why the GDP number avoided negative territory last quarter.
I'll let John Crudele explain it to you.

In coming up with the 0.6 percent annual growth figure, the Commerce Department decided that inflation was just 2.6 percent.

That's lower than Wall Street had been expecting and a tad higher than in the fourth quarter.

Still, the 2.6 percent figure for price increases used for the GDP calculation was nowhere near the 4 percent inflation calculated by other government agencies.

If inflation, for instance, had been 4 percent then the nation's economy would have contracted by an 0.8 percent annualized rate.

And not only that, if inflation was being honestly reported the economy would have contracted in the fourth quarter of 2007 as well.

In other words, there would have been two straight quarterly declines in GDP and the debate over whether or not we are in a recession would be settled.

In the 4th Quarter of 2007 the GDP was also reported to be 0.6%. The government used a 2.4% inflation rate to calculate the GDP of that quarter.
In what universe is the inflation rate only about 2.5%? Even the vastly understated CPI shows more than 4% inflation.

And finally we have today's unemployment numbers.

The U.S. lost fewer jobs than forecast in April, and the unemployment rate dropped, signaling that the economic slowdown may be milder than the 2001 recession.

Anytime the economy loses jobs but the unemployment rate goes down you should be skeptical, and this time is no different. There are a lot of things wrong with this report, so stay with me here.

1) The number of people not in the labor force but still want a job rose from 4.492 million to 4.677 million.

2) The number of part-time workers rose by 661,000, while multiple job holders increased by 131,000.

3) Unpaid family workers (who still get counted as part of the labor force) increased by 26,000.

4) People unemployed for 15 weeks or more went from 2.698 million to 2.911 million. Average duration unemployed went from 16.9 weeks to 18.1 weeks.

5) And finally there is the report that the economy only lost 20,000 jobs last month. The fact is that 267,000 jobs were "created" by the BLS' birth/death model, which is a statistical model used to calculate jobs that don't get picked up in any government survey. In other words, they added in more than a quarter of a million jobs that may or may not exist, just to get a negative number that doesn't look as bad.

And so there you have it. One recession officially avoided by the timely use of truthiness.

Comments

Well ... a little disagreement here ...

First of all, just about everybody including the BLS concedes that birth/death adjustments will miss a downward turning point, and that criticism is well made. One suspects the adjustment will be, uh, re-adjusted after this recession.

Further on the GDP,

Stirling Newberry has a similar, more theoretical take on the GDP number (and BTW, if anybody has his email, it would be terrific if he could be persuaded to crosspost here). Anyway, here's the takeaway quote about the GDP deflator you reference:

You want "two quarters of negative GDP?" Well you've got them measured in the world average of currency. What has happened is a slight of hand to turn the devaluation of the dollar into non-inflation.
How did this work? Well gas prices and housing prices are in tension. The fall in housing prices is being used to offset the rise in gas prices in the GDP deflator. Presto! Having to bleed money to hold on to your house becomes "growth".

Now, I seem to recall that during the housing bubble most of us were excoriating the BLS for not including housing inflation in their CPI. So why should we ignore it now? Again, with reference to Stirling's point, when you substitute the Case-Schiller housing index for owners equivalent rent in CPI, here is what you get:


courtesy of Tim Iacono of The Mess That Greenspan Made

A very interesting graph, no? I rather think that Tim's graph is the most accurate one. Energy and food inflation are the only things keeping us from potentially a deflationary spiral. What say you?

On the government cooking the books, I usually want to stay with the official numbers, on the theory that a watch that is always 10 minutes too slow still gives me very usable information! To wit, here are the GDP and CPI charts from Shadow Government Statistics:

If I want to know if the GDP is going up or down, or if the inflation rate is increasing or decreasing, don't both the official and the SGS charts give me the same information? The direction of the inflation rate, and the GDP growth, is virtually always the same. Granted, in our current case the SGS number would be negative, but is there really a huge difference between +.6 and -.6? I think the economy (as a whole) is likely to improve by Q4. If so, the line on the GDP graph will be pointed upward on either the official or the SGS chart.
If we call for a recession, and then the official numbers don't confirm out prediction, we shouldn't whine about the official numbers not bearing us out. If anything, we should indicate that we are calling for a recession using the SGS formula.

Sorry for the lengthy wonky reply, but isn't that exactly the kind of discussions we want to have here?

You must have Javascript enabled to use this form.

I'm not sure what you are disagreeing about

You sort of showed where I was wrong, and then followed it up by showing how I was right.

I could have gone on a full-blown rant about the CPI, but I wanted to keep this one short.
More importantly, why should the GDP deflator be different from the CPI? And why is the GDP deflator always less than the CPI?
...other than politics, of course.

You must have Javascript enabled to use this form.

Note that the direct story of the report ...

... is that we are certainly sliding into a recession.

If we had economic growth below the break-even growth rate for maintaining employment ... which we had ... and accumulating unplanned inventories ... which we had ... then it is straightforward that we are highly likely to see a decline next quarter.

Obviously the kind of decline that we have seen in the USD will have given a secular boost to employment ... and even with that, the employment picture is still bad ... seasonally adjusted, U6 unemployment is up for two months straight, and three of the past four:
JAN08 9.0
FEB08 8.9
MAR08 9.1
APR08 9.2

... which agrees with the GDP growth rate figure in the neighborhood of 0.6, which would be too low to keep unemployment rates from rising.

You must have Javascript enabled to use this form.

Short take away story ...

... the American people think we are in recession because they are more in touch with labor markets than with the profit side of aggregate income.

Look at U6 unemployment from quarter to quarter:

APR.07-JAN.07: 8.2-8.3=-.1%
JUL.07-APR.07: 8.3-8.2=+.1%
OCT.07-JUL.07: 8.4-8.3=+.1%
JAN.08-OCT.07: 9.0-8.4=+.6%
APR.08-JAN.08: 9.2-9.0=+.2%

While the collapse of the USD from its overinflated value has had a beneficial effect, it was not enough to compensate for the balance of other forces driving unemployment higher.

And what the US public is feeling is the employment-recession, which starts sooner and lasts longer than the GDP recession, because GDP growth needs to be in excess of 0% simply in order to stabilize unemployment.

You must have Javascript enabled to use this form.

That's why it's a "little" disagreement

Good point re why the GDP shouldn't incorporate the CPI. If headline CPI is really accurate, then it ought to be used for GDP, for consistency's sake at least.

At some point (and it may be by the end of this year), food and energy inflation are going to reverse, and if the housing collapse is still going on, Tim Iacono's number is going to go negative (and perhaps so will the GDP deflator) and then it is going to get very interesting.

You must have Javascript enabled to use this form.

But the CPI alone will omit producer price inflation.

There is no such thing as a precise inflation rate. The whole idea of an inflation rate is a single approximate number to measure something that can only be measured precisely by a list of all price increased on all goods and services sold in the economy.

So the inflation rate is always relative to the question. If the question is the overall rate of price increase, a GDP deflator is the best rate to use. If the question is the rate of increase of cost of production, then a producer price index is the best rate to use. If the question is the rate of increase of the cost of living, then a consumer price index is the best rate to use.

You must have Javascript enabled to use this form.

PPI is consistently higher than CPI

So if you found a happy medium between the CPI and PPI then you would still have a GDP deflator that is far higher than what is being used.

As for an accurate measure of inflation, then I would look at the MZM, which is higher yet. After all, rising prices are a symptom of inflation, not the actual thing.

You must have Javascript enabled to use this form.

Not always

It typically only exceeds cpi late in an expansion and into a recession. By the end of a recession, it is typically significantly lower than cpi:

Not sure about Bruce McFadden's point about GDP deflator. I won't claim to know the answer, but if Stirling Newberry's assertion is correct, than the same thing -- home prices -- are counted differently for gdp than cpi. I don't know why that should be the case.

P.S. Thanks Bob Oak for fighting to keep the site up.

You must have Javascript enabled to use this form.

This is all good informative stuff for an amateur like myself...

.....keep it up. I am of the firm opinion that 'economics blogging' is gonna be a big growth sector in the 'sphere. For all the obvious reasons and because the political bloggers have pretty much gone nuts over ID politics....

Which tells us zip about what's coming next.

You must have Javascript enabled to use this form.

Login people!

Anonymous comments are allowed for those drive by commenter's but you all are having a conversation. I strongly suggest either logging in and/or creating an account if you have not.

That way you can track your comments and see when someone has replied and most of all, automatically post. You won't have to deal with the security letter box!

You must have Javascript enabled to use this form.