Monday, August 24, 2015

Is the U.S. Economy in Recession?

In this post I'd like to briefly discuss what the Fed's ZIRP (Zero Interest Rate Policy) has done to corrupt the money numbers and then explain why, in spite of the difficulty of analyzing corrupted numbers, the United States is probably already in a recession, one that started in the 2nd quarter of this year and continues today.

Explaining the Excessive Growth in Currency in Circulation

Anyone who has read the Strong Form of Monetary Policy that I've described in earlier posts might be wondering why currency growth has not reflected the inflation rate recently. Currency growth the past five years has been on the order of 8% per year whereas per the government inflation has been running at less than 2%.

Because a lot of U.S. currency goes overseas, especially in times of international turmoil, some of the excess growth could be explained in that manner. However, ZIRP is also likely to be partly the cause, although not to the extent that it has influenced growth in demand deposits (as will be explained next). When there is no effective cost to carrying cash, people are likely to be indifferent as to the whether to carry an excess of it, for they're not earning anything on the alternatives to cash, including their NOW accounts. Businesses, in particular, would be less sensitive to having well-stocked cash registers.

Yet, most people won't arbitrarily decide to carry wads of cash around just because it's costless, for the simple reason that they might lose it, or have it stolen. That, of course, includes businesses. Then, of course, there's yet another explanation: Perhaps the government inflation numbers are understated, as some fairly reputable sources do, in fact, claim, and that inflation is really running 4 to 5% instead?

Explaining the Excessive Growth in Demand Deposits

This one is far easier. Shorter term savings accounts are earning so little in interest that people are very likely to carry far larger balances in checking accounts today without worrying about the opportunity cost. When $10,000 is likely to earn less than a dollar in a money market fund over a year, why bother? The result has been that demand deposit growth averaged around over 21% per year over the five years ending in February 2015 versus NOW account growth over that time of less than 6%.

When there's no opportunity cost (in the form of lost interest) to holding cash or demand deposits, people, and businesses, simply hold more cash.

Why No One Pays Attention to the Money Numbers Anymore

The favored number of the ancient monetarists (way back in 1979) was M1, which consisted then primarily of currency in circulation plus demand deposits. Five years ago, in February of 2010, M1 was $1,715 billion. This year's February number was $2,995 billion, up at an annual rate of 11.8% over the five years. Monetarists of the past would be expecting a massive inflation rate to have resulted by now.

Very few monetarists at the time focused just on the demand deposit component, but it has gone in the same five years from $458 billion to $1,277 billion, for a 21.6% annualized rate of increase. Note that it has also become a far larger component of M1 in the doing. Under the strong form of monetary policy that I espouse, inflation would be raging by now if that sort of increase had occurred with interest rates not so close to zero percent.

Obviously, then, the money numbers have contained nothing but misleading information since ZIRP became the order of the day. And, in truth, they've contained little information since 1980 when NOW accounts were introduced, as explained in earlier posts.

Finally, Why Might the U.S. Already be in Recession?

While the above discussion explains why neither M1 nor its demand deposit component should contain much guidance for future inflation, that doesn't mean that demand deposits aren't useful for determining whether or not a recession is looming.

Here are the interesting numbers: $1217, $1202, $1204, $1193, $1206. Those are the demand deposit totals, in billions, for February, March, April, May, and June of this year. And the first four weeks of July only averaged $1217 as well, before a huge burst in the last four days of July that brought the total for July to $1,227 billion. Note the virtually zero growth from February to late July following a five year run of growth in excess of 20%.

As explained in earlier posts, while I no longer believe the Fed is capable, under its current operation procedures, of causing such a contraction in the growth rate of demand deposits, that doesn't mean that the contraction itself has no forecasting power.

Indeed, it seems highly unlikely that the virtual cessation in the steady 20%+ growth of demand deposits over the five years since February of 2010 means nothing at all. To the contrary, one needs to ask if the almost five month period of virtually zero growth in demand deposits is a reflection of something, perhaps an economic contraction that has been underway for several months now?

Under the old monetary regime of the 1960's and 70's, such a cessation of demand deposit growth was usually a precursor of an impending recession, particularly if it persisted for six months or so. Back then, however, the Federal Reserve was usually the direct cause of the slowing growth of demand deposits, something that's unlikely today.

Conclusion: Be Careful of the Second Quarter GDP Revisions

On July 30th the government announced that GDP grew at a 2.3% annualized rate, or about 0.6% for the quarter alone. In three days, on August 27th, the first revision will be announced. Don't be surprised if it's a downward revision, perhaps not to negative territory, but downward. By the second or third revision, it could well be a negative number. And, whether or not the 2nd Quarter number is revised downward, the 3rd Quarter GDP number could very well come in at an outright negative number if the sharp falloff of demand deposit growth holds meaning.

So far, the only suggestive evidence of weakness, besides a long topping formation in the stock market that began, perhaps not coincidentally(?), in late February, is a small drop in the last Leading Indicator number and relatively weak retail sales figures for the past few months. If we are in a recession, the news generally lags the event itself and upcoming announcements will confirm it.

Note, by the way, that the Federal Reserve had precious little to do with initiating any recession that might result, and that it also will have precious little to do with ending it, having boxed itself in from a policy perspective as explained in a recent post.


Rod Everson said...

Well, so much for 2nd Quarter GDP being revised downward. It's now 3.3% annualized.

The Third Quarter GDP could still be surprisingly weak, however.

Rod Everson said...

The Fed just passed on raising the fed funds rate this week on the justification that the world economy, and particularly China's economy, concerned them. Here's an interesting article by Peter Schiff on that justification:

Meanwhile, the past week Industrial Production came in weaker than expected, falling 0.4% vs. the 0.2% decline expected. Perhaps the Fed is really more concerned with what they're now seeing shaping up in the U.S. economy, and China was tossed in to avoid having to admit it?

Rod Everson said...

Today, October 1st, the unemployment report came out unusually weak for September, along with a surprisingly large downward revision of 37,000 to the previous August report. (August is notorious for being revised upwards due to end of summer job corrections, per economists.) The 142,000 gain was well below even the most pessimistic projections. In addition, the less-watched Household Survey had total employment falling by 236,000. (Note: It's "less-watched" because it has a much larger error margin than the survey used for the headline report.)

Two additional points:

1. Readers of this post should note that I'm not saying that a fall in demand deposits caused a recession. I'm not even really saying that it ensured that one will occur. I'm just saying that in the distant past a falloff in demand deposit growth presaged a recession by about six months (and those were usually caused by the Fed, by the way) and so the rapid deceleration since February might be a sign of trouble for the economy. In the past, a peak in February would generate weak numbers starting in August or September.

2. Note that my recession post has a date of August 24, 2015, or nearly six weeks ago, which preceded the Fed's announced concern that the global economy bore watching as a reason to forego a modest tightening.

Rod Everson said...

Yesterday we got the report of a relatively weak Retail Sales report for September, with the previous month also revised downward (from a +0.2% to 0.0% change). The +0.1% reported for September was primarily due to strong auto sales and the overall weakness was blamed on weak sales of gasoline in terms of dollar volume. Leaving out both autos and gas station sales, retail sales were unchanged for September, while the consensus was for +0.3%. The problem is that lower gas prices should be enabling higher spending in other consumer sectors; that's not happening.

In my estimation, the pronounced downturn in the growth of demand deposits since February of this year foreshadowed a recession that has already begun.