The Yield Curve, Part 7
This week we look at the probability of a future recession raining on the current economic parade, looking at two indicators which offer insight into our future. I further explore the interest rate environment and the risks associated with Fed policy. There is a lot to cover, and it should be interesting reading.
But first, let me modestly note that my book, Bull's Eye Investing, debuted this week at #13 on the New York Times Miscellaneous List (investment, business and how to books) and #14 on the Wall Street Journal Business books list (right there on section 4, page 6 in today's Journal). Since the book is not officially to be published until next week, I take that to be a good sign. The book is shipping from the on-line stores (I have heard from many of you who have gotten their copies) and is in most bookstores nationwide. If your local store does not have it, kindly suggest they order a few copies.
I received a rather pleasant surprise last week, as Motley Fool writer Matt Richey wrote a rather enthusiastic review of Bull's Eye Investing. Quote Matt: "This is a must-read book for serious students of investing. Broadly speaking, the first 15 chapters are on trends in the stock market and economy, while the final nine chapters offer investment strategies for successfully navigating those trends."
You can read the review and some of his excellent insights for yourself at http://www.fool.com/news/commentary/2004/commentary040423MR.htm.
On a less exalted note, I would refer you to a "review" of my book at Amazon.com by some guy (who of course does not yield forth a name) ranting about how "A republic was unfortunately the shortsighted arrogant plan of our founding fathers" and worse such drivel. Evidently needing a public forum for his political ravings, he neglected to actually mention the book or suggest he had read it. Oh, well. The scary part is that he is somewhere in your neighborhood and has the ability to actually vote.
Before it is (hopefully!) removed, you can read his "review" for your own amusement (and post your own, good or bad), or even better buy the book by going to www.Amazon.com/bullseye.
The Yield Curve, Part 7
I have been writing about the yield curve since the spring of 2000. It is one of the more important keys to the future of the economy. I have been troubled of late by the lack of a "natural" yield curve, as this most reliable of indicators seems to me to be in limbo. This is more than an academic debate, and goes to the heart of the current debate about Fed policy. It also will affect your investment portfolio big time.
Let's rewind the tape. In June of 1996, Fed economists Arturo Estrella and Frederic S. Mishkin of the Federal Reserve Bank of New York, published a quite important paper entitled "The Yield Curve as a Predictor of U.S. Recessions."
They noted: "The yield curve--specifically, the spread between the interest rates on the ten-year Treasury note and the three-month Treasury bill--is a valuable forecasting tool. It is simple to use and significantly outperforms other financial and macroeconomic indicators in predicting recessions two to six quarters ahead."
Essentially, they showed how every US recession in the post-WW2 has been preceded by a negative yield curve. By a negative yield curve we mean that short term rates are higher than long term rate, which is not the natural order of the world. You should be paid more interest for holding longer term bonds and taking more risk. Higher long term rates are referred to as a positive sloping yield curve. (To see a graphic demonstration of the yield curve today you can go to www.bloomberg.com/markets/rates/index.html and scroll down. You will notice the "curve" goes from the lower left to the upper right (or positive). In fact, right now the curve is rather steep.
But sometimes short-term rates rise above long term rates. Estrella and Mishkin demonstrated that the more negative the 90 day average of the difference between the 90 day Treasury rate and the ten year bond rate, the more likely it is that we will have a recession within about four quarters. They assign the odds of a recession with each level of the yield curve, but note that the odds at a particular level might be higher today than in the past for a variety of reasons.
In subsequent papers, they noted that of 26 different indicators, only a negative yield curve was a truly reliable predictor of recessions.
In August of 2000, the 90 day average yield curve was negative and the lowest it had been since 1989, prior to the 1990 recession. Really long time readers know that I became bearish on stocks, as the average drop in the stock market is 43% during recessions.
I wrote in early August: "As of the end of the July, the difference between the 90 day and ten year yield was a negative -.22. On August 2 it was -.27! This is the worst it has been in 11 years. As of August 2, the 90 day average (what the Fed study is based upon) is .19, also the worst in 11 years. Ditto for the 30 day average. It is a negative -.10.
'Now here's the kicker: if the yield curve stays where it is for another 30 days, the 90 day average will go decisively negative. The 90 day average has only been negative one other time in the last 18 years, and that was in 1989, prior to the recession of 1990."
It did in fact go on to become significantly negative. Based upon the Fed study, it was not hard to predict as I did a recession beginning in the summer on 2001, or to begin to call for the Fed to preemptively lower rates at the end of 2000. Since the stock market drops an average of 43% during recessions, it was also not hard to suggest to readers to get out of the stock market. Since rates would be going down in a recession, it was also not hard to suggest that investors load up on long term bonds.
An Artificial Curve?
First off, I will show you a second indicator later that suggests we are nowhere near a recession. So I am not setting you up, gentle reader, to suggest we are. That said, let's look at some problems with the yield curve.
First, the yield curve today is somewhat unnatural. The Fed is holding short rates down to 1%, thus making it nigh on to impossible to see a negative yield curve, even if there were a recession in our future.
Morgan Stanley Chief Economist Stephen Roach thinks the Fed should preemptively raise rates by 2%. Many economists are calling for the Fed to raise rates this week or in June, allowing them to rise to a more "natural" rate of 3-3.5%. Recent Fed governor speeches seem to agree with this line of thinking.
They suggest this as the natural rate because they feel interest rates should be at least 1% above inflation, which in today's environment seems to be about 2% and rising (in terms of CPI, which is a completely different issue).
Today, 10 year rates are 4.51%. Thus, if rates were "natural," when we looked at the Bloomberg site we would see a yield curve which was much flatter.
Before the yield curve goes negative it first becomes flat. Until you see that first flicker of a negative yield curve, you don't need to worry a great deal about a recession in the near future (except for one caused by some surprise "exogenous" event like massive terrorist attacks which could cripple an economy).
Understand, the yield curve is not the cause of a recession. It is a symptom of the confluence of circumstances which precipitate a recession. If it were the cause, the Fed could simply and forever keep the curve positive and we could live in economic nirvana.
In the real world, holding rates down below that natural rate will eventually cause inflation. The market would force rates up at that point, in spite of the Fed. And that is the cause of so much angst among economists and investors. Because they worry that inflation is the near-term problem today, and that the markets are going to react quite negatively if rates are not raised soon, as they will start to smell the odor of this new inflation.
A little inflation, however, is precisely what the Fed wants. I have been writing for quite some time that the Fed will want to see actual evidence of inflation before they start to raise rates. They will want to see that we are completely out of the possible deflation woods. (They will also want to see solid employment market.)
Now, let's visit my concerns. Let's look at the following "what-if" scenario. We have just seen the economy post a slower growth than the consensus expected, but a still very robust 4.2% growth in the first quarter. The inflation signaled within that number was higher than expected, but still nothing about which to be concerned. If we do not see a dramatic rise in new jobs (not the part-time, temporary variety) or inflation posting a significant back to back rise for several months (which does not seem to be the case as yet), the Fed could easily decide to wait until August before deciding to raise rates. At that point, unless the market forces their hand, they might be inclined to wait until the November meeting.
What if I am right and the Fed does not raise rate until November. Let's say inflation does start to creep back over the summer and kick in during late summer, perhaps rising to 3-4%. What would the "natural" short-term rate be? 4%? 5%? 5.5%?
Remember, ten-year rates are 4.5%. Could the "natural" yield curve sometime later this year be negative, suggesting a recession 12 months later? Would we be seeing a "false" positive curve because of the Fed holding rates down and thus assume we are not in danger of a recession within the next 12 months? I could make good cases for both sides of that argument, but the answer is that we would be in completely new territory with little historical precedent.
I tend to think we should look at the natural yield curve, although it is totally conjecture as to what it is. It certainly should be a reflection of inflation.
But whenever the Fed does allow rates to rise, whether it is in June (possible) or after the elections in November (almost certainly), they will do so deliberately. We will be watching that yield curve for clues, as well as the natural rate.
The Potential Economy
Sidebar thought: The independent and respected ECR Research (www.ecrresearch.com) tells us a story I find resonant with my thinking, and they do a good job of explaining the problems. They are based in the Netherlands and Hong Kong, and thus give us a view somewhat outside the fray, but as very interested parties, as what Greenspan does affects the rest of the world. They write this week:
"Under normal circumstances the Fed should already have started a tightening cycle, owing to the strong growth and rising inflation. We indeed expect growth to decelerate later this year. In other words, even if the Fed were to nudge short-term interest rates up this summer, we still expect it to remain deliberately 'behind the curve'. This will eventually lead to higher inflation in the United States in our view.
"However, in our view the question is whether and to what extent the Fed plans to tighten monetary policy. We think it is quite obvious that the current cycle is not 'normal'. The US economy is emerging from a period of deep fears of deflation, and it is generally assumed, not least within the Fed itself, that inflation will not be a problem for some time to come. Conversely, the very low inflation by historical standards means, we believe, that the Fed will want to make sure that it does not stop the economic recovery prematurely. It has declared repeatedly that inflation is 'uncomfortably low' at the moment, and that it wants to see slightly higher inflation as a kind of buffer against deflation. For if inflation remains close to the deflation zone, then the US economy will remain very vulnerable to external shocks and/or another cyclical downswing.
"The financial markets understand that the Fed wants to raise inflation slightly. Even though inflation expectations have gone up, the fed funds and EUR/USD futures contracts indicate that the markets believe that the Fed will only bring short-term interest rates back to neutral (i.e. 3.5-4%) by the end of 2005. In other words, the markets believe that the Fed will continue to stimulate growth over the coming 18 months (albeit gradually less so). In this respect we agree with the financial markets. We believe that the Fed does indeed deliberately want to stay 'behind the curve' and raise interest rates by as little as possible. In addition to aiming for slightly higher inflation, the US central bank is thinking along the following lines in our view: - After the bursting of the bubble of the late 1990s, US economic growth stayed below 'potential' for several years. Consequently overcapacity developed, and this overcapacity is the main reason for the deflation danger in the United States. To eliminate this overcapacity, growth now has to stay above 'potential' for some time...
"However, there is considerable uncertainty about the level of 'potential growth' and hence about the level of overcapacity. Because of the productivity hike since the mid 1990s, it is quite possible that 'potential growth' has moved to a higher level. 'Potential growth' is determined by the increase in the labor force on the one hand and the increase in productivity on the other. It is, as mentioned, the growth rate which an economy can achieve without generating inflationary pressures. Conversely, however, this also means that when actual growth drops below 'potential,' the effect is disinflationary. In other words, although high productivity growth creates more wealth over the longer term (after all, more can be produced with less effort), it is difficult to keep growth above 'potential' in the current situation. It should be borne in mind here that 'potential growth' cannot be measured exactly and is always an estimate. In particular because of the extent to which the productivity hike is structural (i.e. permanent) or cyclical (i.e. temporary). It is likely that the introduction of new machines and technologies has made some of the productivity growth structural, and that hence 'potential growth' is now at a higher level.
The Fed has been singing the praises of the new technologies for some time, a clear sign that it attaches great importance to them. A number of Fed officials have even declared that 'potential growth' may well be around 4.5% at the moment, rather than the generally accepted 3.5% or so. Although this should not be interpreted as the Fed's official position, it is worth remembering that there is uncertainty on this aspect within the central bank.
"In our view this uncertainty also contributes to the Fed's intention not to take any risks in its fight against deflation by curbing growth too quickly. After all, if 'potential growth' is indeed around 4.5%, then even at the consensus view of 4.5% growth this year, it is very doubtful whether the overcapacity can be eliminated. (The Fed also takes the view that if 'potential growth' turns out to be lower, actual growth exceeds it and excessive inflationary pressures are generated, it can nip the latter in the bud quite easily.)"
Sidebar #2: The GDP deflator, which is quite the misnomer, is supposed to be an estimate of inflation. The subtract this number from the nominal GDP to get the real (or inflation adjusted) GDP. Much of the growth in the deflator this month was in the cost of labor. Employment costs accelerated by 1.1% in the first quarter of 2004, well above expectations and on an annual basis something to cause your eyebrows to rise. However, when you look at the details, you find income rose only slightly. Once again we find benefit costs account for a significant portion of the hike, rising 2.4%, while wage cost acceleration is still very timid.
I got my annual bill for increase in the cost of health insurance. I wish employee benefit costs had only risen 2.4% over the last year!
Coincidentally Lagging Indicators
Trading wizard and market guru Dennis Gartman (the Gartman Letter) often writes about an interesting predictor of the direction of the economy. Let me let him explain it in his own words:
"IF WE HAD ONLY ONE INDEX....: We've said it countless times before, and we shall almost certainly say it countless times in the future, but the simpler we keep economic forecasting the better we are at it. Minutia plays well in Ph.D. dissertations, but it proves nearly worthless when it comes to forecasting where the economy shall be six months and a year hence. Thus, we have searched for years for one simple index that serves us well regarding the direction of the US economy and we have held securely to this same index for many, many years. As our long standing clients know perhaps all too well, this index is the Ratio of Coincident to Lagging Indicators, which visually tells us a great deal about the direction of the economy going forward... and tells it more consistently and with less 'noise' than the government's own Index of Leading Economic Indicators, or than the myriad numbers of other indicators that the greatest economic minds whose job it is to do such things can devise.
"Since the late 50's, the Ratio has turned down 10 months, 2 months, 11 months, 10 months, 8 months, 0 months and 10 months prior to the onset of each of the recessions officially 'sanctioned' by the committee responsible for dating each. On average, then, the lead has been just a bit more than 7 months. Let the economic sophisticates take issue with our thesis; we find that it works; works well; works easily and remains uncommonly useful... and until proven otherwise, that is our story and we are stickin' to it.
"The Ratio has been moving steadfastly upward since it turned higher in late '00. Yesterday, with the Index of Coincident Indicators advancing .2% while the Index of Lagging Indicators fell another .1%, the ratio of course moved higher again. Unless something terribly untoward happens in the next several months, we can push the envelope for any possible return to recession out seven months hence... if even then."
When I look at the actual ratio, I find it does indeed turn down prior to recessions. It also turns down prior to slowdowns that do not become recessions, so in that sense it can give some false "positive" signals. But combining that ratio with the yield curve, even if we guesstimate a natural curve, suggest that barring some unknown event, we are not due for a recession this year and probably not until the second quarter of 2005 at the earliest. Neither the yield curve or the ratio looks out more than about a year, so we are left to guess with no basis other than instinct. Bulls will predict a boom and bears will predict recessions. It is a somewhat pointless exercise, as so much can happen within the next 12 months that any accurate prediction based upon today's numbers is pretty much luck.
But the two numbers we looked at today suggest we avoid a recession for the next quarters. I will update you when I see a change.
The Sox are in Town
As I finish up this letter on a Friday night, I look out my office window and see the Boston Red Sox taking batting practice. The game will start soon. Strangely, my Texas Rangers are actually in first place and doing well. This is the first time in ten years we have had a team without at least one, if not three, acknowledged Hall of Famers, or at least none we can suggest today. It will be interesting to see if we can remain in first place after three games with the red hot Red Sox.
Next week, I will look to report to you some of the thoughts of the various speakers at my recent Strategic Investment Conference. Richard Russel, Martin Barnes, Rob Arnott, George Gilder and Scott Burns all gave excellent presentations.
And a final glance shows the field crew putting the tarp on the infield. Evidently they think some rain is headed our way. I better hit the send button and try to beat the downpour.
Your forgot to bring his umbrella analyst,