Investment and Interest Rates
In a recent post, I noted that actual non-residential fixed capital investment doesn’t show the pattern one would expect based on optimizing models at all. Ugh that sentence was convoluted and so is the post it describes. In fact, the puzzling pattern is really very simple. Non residential fixed capital investment (nrfinv) is high when interest rates are high.
This graph Shows an extremely high ratio of nrfinv to GDP in the late 70s and early 80s, exactly when nominal interest rates were highest. It also shows high investment in the late 90s during the dot com bubble/boom and in the early 20th century at the same time as the even more dramatically high levels of residential investment. In particular, the correlation of nrfinv/gdp and Moody’s index of Baa rated corporate bonds (ibaa) is extremely high 0.77 over the whole sample of available data. Over the period 1947-1995 it is an amazing 0.916.
This correlation is strange for two reasons. First the sign is surprising. Other things equal, one would expect high interest rates to cause low investment. note the brick red curve in the graph is the Federal Funds rate — a policy instrument. Second the interest rates are nominal. Sooner or later, I will try to understand what was going on. If you need help in developing an ongoing wealth management plan which is tax-sensitive and reduces risk, click here now.
Another question is: why did I just learn about this pattern ? In 1995 the correlation using all available data was over 90%. Why wasn’t this noted even as a puzzling fact ? I can answer this question. I have been playing with nrfinv/gdp and ibaa off and on for months. I have noted positive coefficients on interest rates. I have thought that they have the wrong sign and must be spurious. I am not as respectful of conventional models as most macroeconomists, but I do reflexively avert my eyes from some summary statistics which are too ugly to contemplate.
Here is another version of the graph a couple of FRED commands away from which shows the strange pattern more clearly.
The high investment rate for non-residential fixed capital investment during periods of high interest rates is perfectly understandable if one thinks about the internal politics inside businesses, dominated as they are by the financial executives. Interest rates in general are high when the economy is humming along on all cylinders. Financial officers sign off on major investment only when they feel comfortable that the economy is “sound.” That means they’ll be looking backward for their proof that their judgement is supported. Only if general business conditions support it (meaning high interest rates) will corporate finance approve those investments. Many businesses may also be maxing out productive capacity and will be investing in a step function upward in production plant, beyond their immediate need for the next 2-3 years but justified by the recent business growth. Throw in the outsize investments in energy exploration and production capacity investment during high economic growth rates with their attendant high interest rates and the counter-intuitive cycle starts to make some sense.
I added yoy % changes for commercial loans and real estate loans to your graph.
The growth rate of loans will plateau as interest rates rise. But loans still increase yoy except after recessions.
It is interesting now that real estate loans are flat. Low interest rates inhibit banks to lend. They receive less profit in the long-run. So many sales over the last couple years were done by cash or equity investment. The return on bank loans would be dismal, so it was better to invest in real estate through equities by private investors. An private investor can share in the eventual capital gain profits, while a bank would not.
“…This graph Shows an extremely high ratio of nrfinv to GDP in the late 70s and early 80s, exactly when nominal interest rates were highest. ”
Also exactly when REAL interest rates were lowest.
Here’s a DeLong graph of real rates :
Rates were historically low in the late 70’s , but this may help explain the early 80’s surge in nonres investment as well , if financing was secured sufficiently in adavnce.
Thanks for the thoughtful comment, but the data are not cooperative. I started looking at real interest rates — in particular the nominal baa rate minus consumer price inflation over the past 12 months I call this lrbaa (the l is for lagged inflation)
I get correlation of 0.57 of nrfinv/gdp and that version of the real rate 0.58 using only data from before 1995).
lraa is Moody’s index of Aaa corporate bonds minus the last 12 months of cpi inflation. the correlation of nrfinv/gdp and lraaa is 0.53.
These versions of the real interest rate were normal in the late 70s and high in the 80s (much higher than ever before or since in the available data set).
Have you looked at the possibility of causation running from residential investment to gdp to nonresidential investment ?
It looks to me like housing starts generally leads gdp , then nonresidential investment follows with a considerable overall lag – a few years (relative to housing starts). Here’s one paper (2001) I found that looks at this :
“….What we have found instead is that capital formation in the residential sector (housing) causes GDP growth, which in turn causes capital formation in the business sector (plant and equipment). ”
The two plunges in real rates shown in DeLong’s graph around ’73 and ’79 correspond nicely to the rising housing starts.
Thanks for the link to the 2001 paper which is very interesting.
On the explanation of my graph, I think two things. First the extremely high nrfinv/gdp occured without extremely high GDP growth and without high capacity utilization in manufacturing. Second residential investment/gdp was about normal soon before the very high nrfinv/gdp and the fluctuations were small. Including the ration of housing investment to GDP lagged one year, or lagged one year and lagged two years does not reduce the coefficient of nrfinvgdp on the baa interest rate (it is very very slightly larger).
Regarding my graph, I think the causation is high non residential investment convinces the Fed that the party is getting out of hand and it is time to take the punch bowl away. So it raises interest rates. It sure seems to me that causation is from non residential investment to interest rates (if it were, one would expect the positive correlation).
It just so happens that I think that housing investment causes GDP growth. I actually had been planning to write on how nrfinv, residential fixed investment and inventory investment are very different and lumping them together is a very bad mistake.
Then I got stuck on how very weird nrfinv has been.
A story about the US business cycles since WWII is as follows
1. except for the last 3 recessions, recessions occur when the Fed decides a recession is needed to fight inflation.
2. The Fed causes a recession by causing a high short term safe nominal interest rate. This causes high mortgage interest rtes (the pattern is clear — why this happens to mortgage rates is not clear).
3. This causes a sharp decline in residential investment and a recession.
4. When the Fed decides that people have suffered enough, it reduces interest rates and the economy recovers.
Note no role for non residential fixed capital investment (the story works if there weren’t any and it works if nrfinv fluctuations are caused by GDP fluctuations which are caused by fluctuations of housing investment.
I will pull your comment to the main blog when I get around to writing the post on different kinds of investment.
Yes , I can see how it may be that the Fed viewed vigorous nonresidential investment as a sign that it was time for “last call”. I also agree that era from Volcker to the GFC represents a significant break or regime change from the rest of the post-WWII period.
Here’s a couple of interesting reviews of Fed policy that describe the “credit rationing” that was in place through most of the earlier period , as well as the changing tolerance of inflation in the 60’s and 70’s. Uncle Milt put a stop to all that , helping to bring us to the happy space we occupy today :
One more link , from the Cleveland Fed this year :
“Private Fixed Investment’s Recovery: Not So Bad After All”
Unless this post is a joke or comes out of sarcasm (economists tend to have a weird sense of humor), I really cant believe this is actually written by a reputable economist. A simple RBC model will give you the result that interest rates and investment are positively correlated. So will a bunch of other models.
Your graph shows that the LM curve is the determining factor. The economy follows the LM curve through a biz cycle. The IS curve is really designed to be a loyal lap-dog for where monetary policy wants to set the LM curve while business expands.
Its like how the aggregate demand curve just follows the GDP point on the aggregate supply curve in a mindless way.
The determining factor for interest rates relies on money demand and money supply as a central bank wants to portray them. Normally the short term real interest rate rises through the expansion of the biz cycle. Where is that real interest rate? It all depends on how a central bank views spare capacity and potential GDP.
A central bank can just keep shifting the LM curve if it wants too and the IS curve just follows. Then the whole ISLM model becomes a self-fulfilling prophecy for the central bank.
Reverse causation? Maybe interest rates follow investment?
When greater investments are needed to meet demand across the economy, demand for investment capital increases and interest rates will rise. I would expect interest rates to follow investments and interest rates to be only a minor factor in the decision to invest.
At the micro-level, a business will invest in expansion when demand is expected to exceed capacity or when efficiency or regulatory considerations require investment.
In the 70s and early 80s large investments in fixed capital were needed to 1) meet the demands of the boomer market, 2) address the rapid rise of energy prices by investment in conservation, fuel switching and process changes 3) move production from the rust belt to the air conditioned, cheap-labor south and 4) meet stricter pollution control and environmental standards by changes in practices that required fixed investment. The changes in fixed investment between 1976 to 1985 are not easily explained by interest rates.
High interest rates are correlated with high inflation rates. This may make future investment more costly even if future interest rates are lower! If investment today is cheaper but at a higher interest rate, the project can be refinanced at a lower rate and repaid from inflated dollars. Note that recessions are correlated with drop in demand and are followed by drop in investment. For these reasons, interest rates may effect the financing of an investment, but have little influence on the decisions to invest and thus have only a minor effect on overall investment.
Investment follows demand. Interest rates are a problem to be solved.
If interest rates are an afterthought in the investment decisions, the interest rates should follow the investment, rather than investment following the interest rate? This would make sense in expanding micro considerations to the macro level and more easily explain the observed correlation.
I am still promising to get around to trying to explain the strange patter. I will certainly pull this comment back. In fact, I don’t have all that much to add.
I thought of some of the same things you discuss (honest I really did). I had forgotten about the investment required for compliance with then new environmental regulations. I recall that this was definitely signficant. I also didn’t consider the rust to sun belt movement.
On the boomers, I’d guess that a lot of the issue is new plant and equipment for the new workers. There was extremely rapid growth in employment during that period. This wasn’t noticed because unemployment ranged from normal to very high and wages stagnated — that is the change due to increased supply of adult workers does not please those workers (also the media stressed the negative in the 70s).
Now the data aren’t totally cooperative on this — correlation of employment growth and investment isn’t super strong. I think the building for boomers story has to be about long term highly predictable shifts (not the wiggles and jiggles).
On inflation, sure investment should be affected by real not nominal interest rates, but, as I mention in a reply above, the correlation of investment with real interest rates is positive too. Here I think there might be something about inflation, the tax code and investment (I am thinking of the lower taxation of capital gains compared to income which encourages re-investment — notably the weird nrfinv/gdp period ends sharply with the 1986 tax reform which made the rates more similar).
I too guessed that the key is capacity utilization, but the data on estimated capacity utilization in manufacturing are not cooperative.
I assume dealing with the huge shifts in the relative cost of energy muszt have been important.
In sum; I definitely think the rust belt to sun belt move is important. nrfinv includes shopping malls and office buildings. I remember what a difference air conditioning made (I’m from Washington DC and old enough to remember the arrival of a central air conditioning unit in our family home). I think the building for boomers must have mattered. I am sure anti-pollution equipment was important *and* measured. Dealing with oil shocks almost had to be important (and timing right).
I think it is clearly true that the interest rate which matters for investment is the mortgage interest rate which affects residential investment. It even makes sense that interest rates have a low effect on equipment investment as they are dwarfed by depreciation rates. I don’t know abou their effect on investment in non residential structures.
well, this may be Simplicio’s contribution to the dialogue:
but wouldn’t it be the case that you have more investment when interest rates are high exactly because those interest rates are telling you something about the expected profitability of investment, and hence the demand for money?
to put it another way, it seems to me the “model” expects people to invest more when money is cheap (rates are low), but the fact may be that people invest more when they expect a higher return and invest less when they expect a lower return, and the interest rates reflect the demand for money.
or did you already say this? [i realize i have absolutely no understanding of how the Fed “sets” interest rates.]
See if this helps:
Chart 1, Ind. Prod.; Cap. and Cap Util
I have no idea if your comment was directed to me or to someone else, but if to me… no, it does not help.
Perhaps I am not smart enough.
But while I am here representing the not-smart-enough constituency, I’ll add to my prior comment that it was my impression the Fed raised interest rates to cool the economy, and lowered them to stimulate it. If so, it would seem to me that you would expect to see high rates during times of high investment and vice versa while the effect was catching up to the cause, if it indeed does.
Without making more of a study of this than i have time and perhaps talent for I would just suppose that my confusion rests on my failure to understand whether Walmann is taking into account the timing and direction of the counter-to-theory correlation he describes.
You snuck in at the same time my comment went up in relation to Robert’s comment: “I too guessed that the key is capacity utilization, but the data on estimated capacity utilization in manufacturing are not cooperative.”
Also I would agree with your analysis of the Fed except I see their triggers arriving too late or too early. In 2000, the Fed could not get down fast enough.
NP it is not a joke. First note these are nominal interest rates — the correlation of investment and real interest rates is lower. Obviously RBC models are uh R and so this bit must be a puzzle.
But second, the way in which one gets high I and r in RBC models is that unusually rapid technological progress causes a high marginal product of capital which causes high I and r.
In fact TFP growth was extraordinarily low during the period of extremely high investment. I guess your comment makes a lot of sense, given the post, to me it is too obvious given the graphs that the late 70s and 1980 data don’t fit any existing model at all.
In any case, the claim that standard models explain or predict or fit the correlation coefficient which I calculated is easy to test. Simulate a standard model 10,000 times and calculate the correlations with the simulated data. I very much doubt that 500 or more will be as high as the empirical correlation coefficient.
That said, your comment is appropriate given the post. I didn’t explain why the graph is so completely unlike anything one would expect given standard models and the discussion of that time period (either when it was happening or since).
Coberly excellent comment. I do think the pattern is due to the Fed raising interest rates to cool the economy. Here the point is the positive correlation with the Federal Funds Rate. The reason this is interesting, is that it is not there in the standard model of the Feds policy in whihc the Federal Funds rate is a function of lagged inflation and unemployment (this is called a Taylor rule).
Now high investment, low unemployment and high inflation are all somewhat correlated. But the correlation is much lower than the huge correlation between nrfinv and nominal interest rates.
I think the pattern can be explained by discussing how the Fed is deciding to cool the economy some times, but it is not at all consistent with the standard assumptions about how the Fed measures the economy’s temperature and decides to cool it (I will write a new post about this).
Just cause you asked what the hell the Fed does, basically it targets the Federal Funds rate and always hits that target,so empirically the Federal Funds rate is whatever the Fed Open Market Committee wants it to be. The Fed actually loans money to banks (technically depository institutions not investment banks but that distinction is moot post 2008). The rate it charges is called the discount rate. Except in late 2008 and early 2009, this interest rate is irrelevant because banks always could (and can again) borrow at a slightly lower rate (the Federal Funds Rate) from each other.
In practice the Fed gets that interest rate to what it wants to be with open market operations in which it buys and sells Treasury bills (also recently when the FF rate is almost exactly zero and the FOMC still wants to stimulate they do other totally huge open market operations called quantitative easing).
Thanks for the response. I accept that my comment had to do with qualitative trends and I havent checked if reasonable models can generate this correlation quantitatively. I gave RBC as an example since its the most popular model out there, but if TFP data does not support it, one can also appeal to models with demand shocks which can get the correlation in that direction. The DSGE literature does support the view that most short run fluctuations are driven by demand shocks.