Sic Semper Tyrannis

Tobin Smith writes in some investment commentary today about the "negative savings rate", and how it is a myth. Smith, who I often respect for some of his investment and economic calls, has a huge blind spot for anything that might reflect negatively on the US economy overall. The negative savings rate is an instance of this blind spot, as his remarks today indicate:

Follow up:

If you listened to the radical left or radical right, you would think our economy was about to run out of "capital" because our "national savings rate" was so darned low.

That's right folks, you have to be on the "radical left" or "radical right" to think the United States has a savings problem. Never mind that most of the "center" of this society (economically speaking) is the source of this negative savings---and they certainly know it.

Well, the state of our nation's "capital" was a LOT better off than some people could fathom, especially due to the insane, arcane way we measure national savings.

I've been telling you that for a while now. Now I have proof.

Who can argue with proof? Especially against insane, arcane things.

[Ed. note: Of course, anything goes as far as "proof" is concerned in economics.]

In a preliminary study released last week, the Bureau of Economic Analysis said that research and development investment, or R&amp, accounted for 4.5% of growth in real GDP between 1959 and 2002.

How big is that? Investment in bricks and mortar was just above 2%.

Here's the insane part. Currently the hundreds of billions of dollars that businesses spend each year on research and development (R&amp) are not counted toward economic growth -- i.e. they are NOT an investment; instead, they are an expense.

Oh no! They aren't?!

Well, actually, thats not a bad idea. You see, while Tobin Smith has never worked in research and development, I've always worked in research and development, and I've noticed something quite glaring about it:

It usually fails.

That is, most dollars spent lead to little or no "value" created. This makes sense, as many ideas are no good, or fail to be received by the market, or simply are ahead of their time in terms of feasible implementation.

In fact, I'd be surprised if even one in ten research and development dollars led to an eventual recouping of that same dollar. The record is even worse for public-originating R&amp.

I'm not saying that R&amp isn't profitable; on the contrary, it is critical and drives the majority of economic progress. But you simply cannot account for R&amp by chalking up each dollar spent on it to a dollar "saved", as if it was in some sort of deterministic "investment" like a bond.

That money, once spent, is as good as gone.

Instead, R&amp is properly accounted for by counting the up-front cost as spending, and profits from successful developments as earnings. There simply is no direct way to tie the two together; while management must be aware that R&amp leads to future profits, any honest accountant would laugh you out of the office if you tried to argue an R&amp dollar was an "investment" in the accounting sense.

Further, if you count the "invested" R&amp money as savings, as well as the profits from successful development as earnings, you've just counted the same money twice. How does that make any sense? The illicit, underlying manouver here is to take the fundamentally non-balance sheet phenomenon of innovation and pretend it can be precisely accounted for. This is totally bunk.

If you don't agree with me, I have plenty of R&amp "investment ventures" I can sell to you. The payoff could be very high (say 100:1); I'm sure you don't mind if the probability of success is only 1 in 100. Since we're calling it an "investment", after all.

Smith's comparison to bricks-and-mortar is also specious; each dollar of physical plant is a dollar immediately received in use-value. R&amp spending doesn't have that kind of reliability. (Note here also that depreciation is essentially an accounting compromise between the certainty of high initial "capital value" of spending on physical plant, and the near-certain low eventual value).

Let's see what else Smith has to say...

Say you are a company that has $1 billion of sales and spends $100 million on research and development projects. With the current approach to accounting, this investment in people, equipment and everything else is an "intermediate" expense. That makes the expense deductible off one's tax bill.

But treating R&amp as an investment would make the U.S. economy 3% BIGGER and our "national savings rate" 2%-3% HIGHER.

Ahhh, so now we see what this is really about: we need to go through these semantic contortions specifically to try to salvage the US's flagging balance sheet!

The implicit argument here is that if we re-calibrated the savings rate by adding to what we count as savings, whatever problems critics are pointing out with the "negative" savings rate would vanish.

But critics aren't just pointing out the negative savings rate: they're pointing out a number of connected ills, such as declining real incomes, enormous levels of overseas borrowing, a high rate of credit-fueled spending (which is exhausting--an additional problem), and finance bubbles.

Even if we re-scaled the savings rate, shifting it up by 3% so that the national savings rate was now in the vicinity of zero, these problems would remain. This should be pretty obvious: all we will have changed would be the accounting; all of the ills connected to the savings phenomenon would still remain.

Thus I think it is appropriate to inaugurate a new AutoDogmatic Fallacy: The Fallacy of Statistical Deus Ex Machina. We define this fallacy as one in which a specific measurement is illicity substituted for some (typically broad) empirical claim, and then it is argued that the measurement itself is simply wrong in degree. The measurement is substituted for one that produces favorable numeric results, but this is almost totally irrelevant to the initial claim or observation.

Anyway, back to a couple of Smith's other points:

He initially suggests that the argument of the bearish crowd is that we are "about to run out of capital". But I've never heard of such a ridiculous claim; this is a straw man.

If he means money, then this is trivially false: the government can always print more money in a fiat money system.

To be more generous, perhaps he means capital in the sense of surplus, investable wealth (independent of money). Then, it would be more correct to say that "the bears" argue that we are at risk of running out of wealth, except for one small detail:

We've already run out of wealth. That's what the negative savings rate means---and failing this, the massive borrowing from overseas, extension of consumer credit, and inflation of the monetary base.

So only a fool would argue that we "may" run out of capital, and only an economic flat-earther would argue that we haven't already.

Moving on to another point, Smith's deus ex machina, accounting-based salvaging of the savings rate ignores the standing of the US relative to other countries. It is not just our Asian trading partners that have extremely high savings rates (and even those countries are not full of people who are plowing most of that money directly into the US economy); all Western countries have savings rates dramatically in excess of the US. That money is not all going to the US; not by a longshot.

So the resolution to this savings conundrum is not that the US is "really... really awesome", but that the US doesn't save because its balance sheet is in tatters, while basically all other major countries have plenty left over to invest both in and amongst themselves.

Further:

In most countries, the return on equity capital at the corporate level is very low -- lower than what investors can get in the risk-free bond market. The best example is Japan, with an average return on equity at the corporate level of less than 3%.

In the U.S. we average well above 8% -- it's why we are capital importers rather than exporters -- capital invested in U.S. companies returns a higher rate of growth.

Close but no cigar, my silver-spooned friend: inflation in Japan is zero, whereas its 4-7% in the US, so return on corporate capital in the two countries are approximately equal and probably better in Japan. (On a related note, the real historic return of the Dow is 1.6%).

Further, it is actually nonsensical to claim that being a desireable source of investment makes a country necessarily a capital importer: if foreigners put in capital to invest it, they will eventually take out even more capital to retrieve their capital gains. Duh.

Besides, if this were true, why has the dollar been falling on the international exchange since 2002 (or even more broadly, since the 1970s)?

Finally, if it were really true all this (trade deficit) money was going to productive investment in the US, then it would all go into the private capital markets. But it isn't... more than $6 trillion of it has gone into the government in the past half decade---$4 trillion to the federal government alone.

I find it hard to believe that's a healthy phenomenon.

Note: Here's another article in the spirit of Smith's arguments. In it you'll find implicit claims such as one whereby the US's lead on Japan in education+R&amp spending by approximately 3% of GDP makes up for a 25% of GDP differential in saving. I'm all for accounting for previously-neglected factors, but damn, learn some subtraction, people.