The choice channel for low real interest rates

There is a new NBER working paper by Iachan, Nenov, and Simsek on the question of why short-term real rates are so low.  Their model is simple: the cost of financial intermediation is falling, and furthermore portfolio customization is much easier these days.  Investors can buy many more distinct assets, and so they are more likely to find favorites.  In general portfolio investments go up and asset prices go up, but yields on the low-risk assets go down.  The more you can customize, the more assets you wish to buy, and as part of general portfolio offset/construction, the more of the risk-free asset you wish to hold as part of your efficient diversification strategy.

I think of it this way: the more easily you are able to collect beloved artworks, the less risk-sharing you are doing, and the more you will invest in some kinds of safety as a risk offset.

Pretty neat, we’ll see how it stands up (does it predict the behavior of real investment as well? and what does it imply for the predictability of returns?).

The authors, by the way, are claiming only that this is a contributing factor to low rates on low-risk assets, not the entire answer.  And note that in their model, unlike in standard accounts, falling risk need not cause the short-run rate to rise again.  Their model also predicts that falling securitization leads people to invest more in the very safest assets, causing those yields to decline further; that is consistent with the post-2008 world.

Greater market participation, taken alone, does decrease the risk-free rate in the model.  And arguably greater market participation came along in the early 1980s, more or less when risk-free rates started to fall.

So how do you get out of a liquidity trap?  Well, if the trap is not benign, either restrict portfolio customization or facilitate securitization, in other words enabling a greater number of secure intermediate assets.  Who would have thought?

File under “Don’t think you understand the real interest rate.


Aren't interest rates set by central banks? The Fed is like the control room for a nuclear power plant, with many dials to watch, but unlike a nuclear power plant they only have one knob to adjust to control the process. That knob is interest rates, and it's pretty much jammed against the lower limit of its range.

Interest rates are set by the intersection of the supply and demand curves. While the central banks have shifted the supply curve toward increased supply, the demand curve has either not moved for three decades or has shifted to less demand.

This can be seen in the velocity of money.

Think of increased supply of corn but stagnant or declining demand, leading to both lower prices and the shipments of corn falling leading to increasing stockpiles of corn.

That's basically what's happening with money. Its piling up with no demand for it to meet the needs of the economy based on declining consumer cash flow and declining investor interest in building new assets which would increase consumer cash flow.

The constant policy demands for lower prices for everything: food, energy, even civilization, means labor incomes are lowered, and lowered labor incomes force lowered consumption spending. With reduced demand for GDP, investing to increase GDP will drive down profits which are the only thing of all the things in the economy that are being forced higher by many public policies from the Wall Street elites to boards of directors, conservatives in politics.

Note that consumers for various reasons borrow money for consumption, which is idiocy, but that is an activity promoted by Wall Street rent seekers - they borrow at 2% and lend at 18-35%, which in three years eliminates all demand for GDP by the laborer who sinks into that hole, at least until he invokes the Constitution wealth redistribution clause taken from religious tradition of debt forgiveness in sabath years - also the basis for sabbaticals for tenured professors.

+1 to mulp. I think Sumner and his minions would disagree with this "demand oriented" argument which correctly minimizes the Fed Reserve's influence (as indeed no other than Ben S. Bernanke found in his 2003 FAVAR paper, showing using econometrics Fed policy shocks only affect economic variables by 3.2% to 13.2%). But to me it makes the most sense, and incorporates TC's "Great Stagnation" theme.

Holy shit a coherent post from the mulp! The last para got a bit shaky, but still, B+!

No, CBs attempt to alter interest rates to achieve targets. Rates are set by the markets.

Also, Fed intervention is not limited to interest rates, they can bid on every asset in existence.

I like this:

"To understand the intuition, imagine Önancial assets as a forest that contains several trees.
The trees could be a metaphor for individual stocks, industries, or mutual funds with di§erent
strategies or styles. Suppose each investor has a higher valuation for certain trees than the
average investor. In our model, this happens because investors have heterogeneous beliefs.
However, heterogeneous valuations of this type could also emerge for other reasons (e.g.,
institutional restrictions to hold certain trees). Suppose increased customization enables
investors to trade individual trees as opposed to buying or selling claims on the forest. As
a response, investors would expand their investments in the trees they like the most, while
reducing their positions in the trees they like less. Moreover, for every (relatively) optimistic
investor that would buy a particular tree, there would be a pessimistic investor that would
sell that tree. Consequently, investors would collectively like the forest more, in view of the
choice channel, but the relative appeal of individual trees would remain unchanged. We
show that this logic is quite general, and implies that greater customization increases the
valuation (and reduces the expected return) of each tree in tandem."

But it does not at all address why short safe assets have historically low returns, while other assets including long safe assets have historically fairly ordinary returns.

It is a bit hand wavey, saying that the natural rate is zero or less and remove all the arbitrage opportunities and frictions it will be zero.

Money has no inherent worth and hence worth nothing to rent. That is what zero rates tell us. One day it will sink in. Then what?

Why would more information induce investors to choose less risk? More information should offer greater choice along an array of risk (and rates of return). Does all this information make investors risk averse? Or does recent experience make investors risk averse? Or do expectations about the future make investors risk averse? This working paper tells us a lot less than it purports to tell us about investors and risk. One simple observation about human behavior (investors are human, aren't they?) is that people expect tomorrow to be just like today, only more so. It's open enrollment season so I will link to a post by Austin Frakt in The Upshot., in which he states that there is a bias toward the plan with a lower premium and higher deductible even when a plan with a higher premium and lower deductible is more efficient (i.e., the total "cost" is lower). Are insurers phishing for phools by offering "choice" (i.e., inducing exchange customers to pick the plan with a lower premium because the overall "medical loss" to the insurer will be lower)? Of course, the bias can be explained, at least in part, by the insured's belief that tomorrow will be just like today. Similarly, the investor's bias toward less risk (and lower return) can be explained, at least in part, by the investor's belief that tomorrow will be just like today. And for the insured and the investor, all is not lost if tomorrow turns out different from today: for the investor, she can switch investments in almost an instant, and for the insured, there's always the next enrollment period.

I might believe that this is a reason TIPS (a thin market of interest only to those with broader portfolios) do not really represent inflation expectations, but I find it hard to believe as a big factor in the broader range of interest bearing investments (held by many people without broader portfolios).

Should we really jump to the conclusion that there's something that specifically needs to be explained about short-term risk-free rates? It seems to me the most natural conclusion is that rates are down across the board, and given that short-term risk-free rates are typically the lowest rates, it's most obvious there. But I don't really see any evidence that the spread between short- and long-term rates or the spread between risky and risk-free rates are unusually elevated. Looking at the historical spread between say BBB yields and risk-free yields, or say the spread between the 2yr and the 10yr (or 3mo and 2yr or whatever part of the term structure you want to focus on), they are not unusual by historical standards. So this seems like an answer going in search of a question.

Fantastically interesting thoughts, as always.

You don't need to model how a population will behave when you know how they behave: old people are more risk averse. At the same time, fewer young people leads to less demand for capital. Look at China, it rapidly advanced into the high debt/ low expected return camp as the effects of the one-child policy hit.

Retired people are taking on more risk today because they need higher returns, not because they want the risk. There's a mismatch in many portfolios and pension funds are in the same boat.

Look at the margin, where there are retired people in need of income and pension funds with big liabilities, both needing high returns. They drive down safe assets and then start driving down returns on riskier assets as they try to achieve higher returns. Meanwhile, the demand for debt is low due to demographics and already high debt levels. The "savers" are desperate to hit their goals so they are forced to save more to make up for low returns, driving up asset prices further.

Their argument seems counter-intuitive to me. If it is easier to invest in things rather than to deposit it in financial institutions, wouldn't this dry-up the supply of money available to be lent?

My thinking is that the lack of profitable opportunities has made the demand (and therefore) the price of money to be low.

It does not seem reasonable that ease of investing could change enough to produce any significant effect.

Comments for this post are closed