Harry Dent’s Demography-Based Economic Theory
[On November 29, 2000, as the U.S. economy was approaching a recession, I sent out a letter to my sister Carol and to a number of friends in which I said in part:
Next year... I expect the [current] economic weakness to continue and to intensify, quite possibly leading to an outright recession. Whether things will develop beyond that, this time around, is impossible to say at present. My understanding of political economy leads me to predict, however, that sooner or later one of these recessions will be severe enough to cause the credit bubble to collapse, thus worsening the situation qualitatively. At that point a long, drawn-out, episodic (ups-and-downs within the downturn) depression will commence. Of course I have been claiming this for a couple decades now, and so far it hasn’t happened!
On December 1, 2000, my friend Kevin responded as follows:
OK Scott.
I’ll go out on a limb and saw it off. The next depression will occur around 2025 with signs of its coming starting to appear around 2017.
Why do I think this is the time line? Because the main driver of economic activity is the demographics of the US population. The main influencer of economic trends in the US are the 76 million baby boomers and they begin to retire or enter middle age in this time period.
As a population ages its consumption patterns will change. This has been shown in a number of studies by various academic economists. The phases are household formation during 20s and early 30s, child rearing during late 30s and early 40s, saving for retirement late 40s through 50s, retirement late 50s through 60s.
We can expect a net buying/investing boom over the next decade with bumps along the way. If you study the flows into mutual funds this year you will see that the flows haven’t really reduced dramatically. This is being driven by retirement plans. Once the boomers get closer to retirement you can expect a move of retirement assets into fixed income and lower risk investments. This will mean a reduction of stock market investing and a falling stock market. Look at Japan as an example of what will happen to the US. They are ahead of us on the demographic curve.
Kevin
Kevin had mentioned before that Harry Dent had originated much of this theory. I therefore replied with the following discussion of Harry Dent and his theory. —Scott H.]
December 2, 2000
Hi Kevin, and everybody—
You have mentioned before that you are a believer in Harry Dent’s economic theory, that the key determinant of the economy over the long run is demographics. I won’t attempt an exhaustive refutation of that theory, but I’ll raise some points for your consideration.
First, for the benefit of others receiving a copy of this letter, who is Harry Dent? He is a bourgeois economist and stock market guru who has made quite a name for himself among investors—for the very good reason that he says what people want to hear. What investors want to hear is that everything is working out great, that the economy and the stock market are booming and will continue to boom, at least over the next 8 years at a minimum. Obligingly, some of Harry Dent’s books are entitled: The Great Boom Ahead (1992), The Roaring 2000s, and The Roaring 2000s Investor.
One interviewer introduced him as follows: “Harry Dent is widely regarded as one of the world’s most accurate prognosticators of the economic, technological and demographic trends in society.” Quite a build up! But what is his actual record? It is true that he predicted a boom in his 1992 book (just as the U.S. was already coming out of a recession!); that’s what his reputation really rests on. It is said that he correctly predicted that the ongoing, prolonged Japanese recession would continue, on the basis of demographic trends (the extremely low birth rate there). But demographic trends also led him to predict that the rest of Asia would continue to boom, and he thus completely missed the severe 1997-98 Asian crisis (which still continues to some degree, and is even showing signs of staging a “round 2” soon). If you go to his web site (at www.hsdent.com) and check out his most recently posted newsletter (May 2000), you’ll find him claiming that the NASDAQ [stock market index] had probably bottomed out, and most likely would soon be heading up again, probably rallying to new highs by the end of this year. (And he continued to predict that the NASDAQ index will hit at least 30,000 and possibly 45,000 by 2008.) That was his May issue. However, on June 2nd, Business Week ran an article: “Harry Dent: A Long-Time Bull Turns Short-Term Bear”. There he said that there was at least a 50-50 probability that the NASDAQ index would falter again over the short term. In other words, he waffles as the situation dictates (like all bourgeois economists).
And if you read his short term predictions in his newsletters or interviews you see very clearly that he tries not to say anything at all definite. Of course, long-term demographic trends can’t tell you much of anything about short-term economic trends. So perhaps we shouldn’t dwell too much on his short-term failures in evaluating his theory that demography is the key to economics. But on the other hand, that means that whatever short-term prognosticating successes he has had are also irrelevant to evaluating the demography theory. In short, he has put forward a theory relevant to the long-term, and made long-term economic predictions on that basis. And the long-term results are not yet in.
So to evaluate Dent’s demography theory we can not yet rely on empirical evidence accumulated since the theory was propounded. We are stuck with:
1) evaluating the historical evidence in support of the theory,
2) looking for logical and coherence problems in the theory itself, and
3) comparing the theory to competing economic theories to see which best explains the facts, and seems most plausible.I’m not going to try to do much along any of these lines here; just bring out a few points.
I won’t attempt the third thing here at all, since that is a very big task, involving the presentation of alternative theories and detailed comparisons of them. I’ll just say that as a proponent of one of the variations of Marx’s crisis theory, Dent’s demography theory looks extremely superficial and simple-minded in comparison.
But what about method 1), looking at the historical evidence? Does it back up Dent’s theory as he says it does? One of the things you have to be very careful about in science is the adoption of a theory based on “retrodiction”—the ability of the theory to “predict” what has already happened. The reason for this is that after the outcome is known it is all too easy to arrange the earlier data to make the theory seem to have been verified. If the data in one area (one country, say, for Dent’s theory) looks “correct”, then this is the data likely to be mentioned. If the data in another area (country) makes the theory look suspect, then this is the data likely to be “neglected”. If the data can be massaged somehow to make things come out “right” (i.e., as they actually did in reality), then it is amazing how “sensible” that sort of massaging will come to seem to the theorist and his/her adherents.
Dent’s theory says that the birth rate affects the economy. But what is actually far more clearly shown in history is that the economy affects the birth rate. During the depression [of the 1930s], for example, the birth rate fell catastrophically in all advanced capitalist countries. A good case could be made for Dent having simply confused cause and effect here!
What about method 2), then—looking for difficulties in the theory itself? Here again you’ve got to look carefully at what Dent is really doing. Let me give one example:
On his web site he has one little page of theory called “The Spending Wave”. He begins by saying: “One simple forecasting tool predicts, with uncanny accuracy, the health of the economy and the stock markets over many decades to come: the Spending Wave.” He claims that the “Spending Wave” predicts the health of our economy by lagging the birth index by 46.5 years—which is the age, he says, when we on the average hit the peak of our spending. (Has that age been constant throughout history, and in all countries? Of course not! This is just one of the many little theoretical difficulties he slides over.) Dent illustrates this “Spending Wave” concept with a graph where he superimposes the Dow Jones stock index curve over the immigration-adjusted birth curve. After massaging the data to make the two curves fit, amazingly enough—they fit! And since they fit for those years, he feels justified in extrapolating out for 46.5 more years. Voila! The Future is made transparent! And the Dow is projected to rise to 35,000 around 2008.
But let’s look at a couple points about how the data is massaged here. First of all, we are dealing with two graphs superimposed. The units at the bottom are years, which are of constant length on both graphs and so can be reasonably superimposed. He moves one graph to the right by 46.5 year units, but we won’t object to that. But the vertical scales of the two graphs are entirely different and unrelated (millions of births in one case, Dow Jones index in the other case.) He has obviously just chosen units and spacing here to make the graphs superimpose correctly! Moreover, to get this to work, he even had to use a logarithmic scale for the Dow! Since the vertical scales in fact bear an arbitrary relationship to each other, the apparent coincidence of the two curves is a complete statistical lie! At the very most you can only say that the directions (up or down) of the two curves tend to agree (in this case and with this limited data).
Well, I was going to go into things further, but I have run out of time. I have to get ready and head off to our Science Book Club meeting tonight—and I don’t want to miss that just to go on at further length against Harry Dent! But, Kevin, if you (or anybody else) would like to respond to what I’ve said so far, or raise other issues related to the demography theory, please do so!
—Scott
Marxist Political Economy Home Page
MASSLINE.ORG Home Page