In the absence of economic rents, the return on corporate capital should generally follow the path of interest rates, which reflect the prevailing return to capital in the economy. But over the past three decades, the return to productive capital generally has risen, despite the large decline in yields on government bonds.
Pointer from Mark Thoma.
For a moment, think that there is just one interest rate. If “the” interest rate is low, then the rate of return on new capital ought to be low. Otherwise, firms would borrow at the low interest rate in order to purchase new capital.
One possibility is that the marginal return on new capital is low, but the returns on existing capital are high. That would be true in an economy where there are economic rents available, due to monopoly power and/or government favoritism. I gather that this is the story that Furman thinks is right.
I would note that there is more than one interest rate. It could be that there is a high interest rate charged to firms that are trying to invest in new capital at the margin. Microsoft can borrow at a low interest rate, but when it buys Linked-In that is not new capital investment.
Despite that possibility, my inclination is to believe that Furman is onto something. Read his whole essay.
It is entirely possible that in a complex economy, larger surviving firms are inherently more efficient entities without the presence of monopoly power or government favoritism.
Deloite’s “Shift Index” contains information that appears to contradict Furman’s views.
While Return on Assets may differ from the “Capital” Furman identifies, it does seem to match (correlate with) other factors in what we refer to as our “economy.”
From the inception (see, Berle & Means) of the current Managerial period of “Capitalism,” (continuing accumulations of surpluses) with increasingly indirect beneficial ownerships of enterprises, there has been a steady increase in Managerial sequestration of surpluses.
It is interesting how, at the ending (or trending out) of each phase of “Capitalism,” the commercial, the industrial and the financial, the ‘villains’ or scape-goats of the disruptions of the transitions are never seen as those coming into dominant positions (such as the Managerial Class – which now also dominates politics and governmental mechanisms).
That is not to say the transition is “bad” or any class is dysfunctional; just that the overall “picture” is continuing to change, and that some attention might be given to the “shift” to a Managed Society.
I majored in finance in undergrad and am not a macroeconomist. That said, I’m always confused by macroeconomists’ notion that returns to capital are the same thing as interest. In finance, the return to capital is the rental rate and interest is the time value of money. Interest is used to discount the return to capital to determine the value of the capital asset. They are distinct concepts.
What am I missing?
What **may** be “missing” is consideration of “interest” as a pricing mechanism in the majority of transactions requiring “credit” or “time value” of asset usages.
When any pricing mechanism is disrupted by factors (political, dirigisme and others) not intrinsic to the transactions (wage-price controls, e.g.) its functions are usually (historically) distorted.
We are currently observing a “shift” in central bank policies from interest *rate* (credit) focus to manipulating the effects of variations in composition of “time value” assets comprising their (so-called) balance sheets (short term – longer term).
The usual problem is the most profitable businesses have little need for capital while the businesses that have a high need for capital are not the most profitable, with the margin set by businesses that can use some capital but are only limitedly profitable.
Perhaps you speak to requirements of externally supplied *additional* “capital.” A successful enterprise produces surpluses, which constitute “Capital.” How those surpluses are deployed – or – if they are simply sequestered by managements is a major consideration in evaluating the continuing *real* growth of an enterprise.
Where are the surpluses (profits) being re-deployed currently?
Are there political and social obstacles to redeployments?
Who is making the determinations; what are their motivations?
Why is there no mention of risk? Yes, return on investment ought to be affected by interest rates, but also by risk. Take the CAPM. R = a + bX, where a is the risk free interest rate, b is beta (risk), and X is the (common to all assets) premium for loading on risk. Then, if interest rates are low but return on investment is high, perhaps firms are investing in higher-risk (higher beta) assets?
I’ve seen this topic addressed many times in one form or another, and I groan each time I see it explained. As I recall, Piketty attempted to address this, and I found his explanation to be crude and lacking. With all due respect, I think Furman is way off base. I think it’s necessary to make the following distinctions in order to address the broader issue.
1. Many of these discussions do not distinguish between the portion of returns to capital coming from cash flow, versus those coming from capital appreciation. This is an important distinction. For reasons described below, capital appreciation can be very positive in a declining interest rate environment.
2. Many of the discussions do not properly distinguish between ex ante and ex post returns. For example, when one prices an asset or security, one makes certain assumptions regarding cash flows, interest rates, etc. With the exception of the cash flows portion, most inputs are assumed to be fixed. In reality, most of these assumptions, including interest rates, change over time as the period being “projected” turns into “actual” results.
3. Another distinction is conflating interest rates, versus changes in interest rates. The key to answering this conundrum between higher returns and lower interest rates is not that interest rates today are low today. Rather, it is that the rates have declined, or may decline in the future. This is an important distinction. In Newtonian physics where distance is a dimension, velocity is the first derivative of distance and acceleration is the second derivative of distance. However, distance, velocity and acceleration are not the same thing. Similarly, interest rates and changes in interest rates are not synonymous.
4. It is also important to note that changes in interest rates affect both positive and negative cash flows, and I’m not sure that this analysis reflects this distinction. As a finance professional, I have an appreciation for this because I typically seek to price equity stakes of certain assets, not the asset itself. Because most assets are partially financed with debt, changes in interest rates can affect an equity holders cash inflows, as well as cash outflows made to finance debt. Also, because debt has a fixed, and often short-term duratiom whereas equity typically has a perpetual duration, there is a duration mismatch between the two that can cause the value of equity to increase during a declining interest rate environment.
Here is how I see this in practice. Say I’m seeking to price the value of a piece of equity using a fundamental approach (as opposed to looking at comparables say), such as a discounted cash flow model (DCF). Say I ran a DCF in a moderate interest rate, moderate ROA environment. I then buy the asset. Over the next five years the economy shifts towards a low interest rate, low ROA environment. The net effect on my return will depend on how the cash flows on my asset changes, changes to my borrowing costs, changes to the exit value and changes to the risk premium (see Jack PQ’s excellent comment). Which will win out depends on the circumstances, but if we were to use a formula in which case the real interest rate was part of the DCF model, that real rate could be expressed as the summation of some sort of constant, plus some factor that relates to the real growth in the economy, plus some other exogenous factor (perhaps a risk premium, see Jack PQ comment). In an environment in which the real interest rate is declining, and specifically is declining because of a decline in this exogenous factor (and thus not wholly attributable to declines in the real growth rate in the economy), then the decline in my cost of borrowing over time will outweigh the changes to my cash flows over time, and thus the value of my equity would increase over time. This doesn’t even address how such changes would effect the “terminal” or “exit” value, but that is a whole other discussion. In summary, this is how one can have a high return on capital in a low interest rate environment. But again, I would not state the situation this way. I would describe the situation more accurately as a high return on capital (including the returns due to appreciating values) in an environment in which the interest rate has declined.
I should add that in such an environment, this most definitely does not imply that returns to capital will stay high. If one believes in a mean reversion view of the world, the opposite could be inferred, mainly that appreciation will slow or be negative in the future, and thus future returns on capital will decline or go negative.
Wow, that was a lot of words to say “but interest rates.”! Interest rates don’t affect ROA. ROA ignores financing.
Bingo.
If an enterprise accumulates surpluses that are not redeployed to expand returns on *those * surpluses, assets increase whilst returns remain flat (or do not increase at the rate of increase in assets). Hence the steady decline in ROA noted in the “Shift Index,” which is continuing.
Good point. I tried to describe some other drivers as well, but you put your finger on the main one.
An argument for ‘one interest rate’ is from arbitrage pricing theory + EMH. If “expected, risk adjusted, net yields on capital assets” are not the same in the short term, then someone (and/or some algorithm) is going to notice, exploit the profit opportunity soon, bid prices up and down, and reallocate to erase the differences. Yes, there are fluctuations, uncertainties, surprises, and all kinds of new information, but any gaps or differences that suddenly emerge shouldn’t persist for long. At least for the easily measurable stuff.
Now, it could be that since an increasingly important type and larger share of capital these days is organizational capital – which is fuzzy and social and fragile and really difficult to observe, assess, and measure – that ‘return per assets’ is now a lousy metric, because the accounting definition of assets is inadequate to capture the reality.
A metaphor could be with gravity. Galaxies spin too fast for theory, so there is a problem. Is it with gravity (interest rates) or mass we can’t see (firm assets)?
With the galaxies, the conclusion was no, gravitational theory and general relativity (APT/EMH in the metaphor) is fine, so there must be dark matter out there.
So I guess that the ‘dark matter’ mystery is on the assets side, not the interest side. It is probably better then to assume one rate as above, and then use that to infer the true market assessment of underlying assets. Subtracting ‘visible assets’ (the ‘hard’ stuff in the accounting / ordinary finance definition) gives one an estimate for the dark matter.
If the software company is the exemplary kind of new company, software companies have a) few fixed assets and b) no way to expand their business beyond hiring more programmers and sales people and c) little competition within their niche and d) locked-in customers. In other words, high returns, few investment opps and few competitive threats. That’s basically what Furman is describining, a world of high returns and low capex and little competition.
“the path of interest rates, which reflect the prevailing return to capital in the economy”
Talk about starting with a mistake and winding up in bedlam.