Uncovering the mysteries of persistently low interest rates

Posted by Alex Frey (@alexhfrey )

Note about the Future of IvyVest 

A couple technical problems with our payment system has brought home the time investment required to maintain a modern ecommerce site. Competing demands on my time have forced me to conclude that I no longer want to make that investment. As a result, I have decided to shut down IvyVest as a subscription service, effective as of the end of this year (Dec 31, 2019). All subscribers in current standing will be issued a pro-rata refund as of this date. It is important to note that this is not related at all to any concerns about the validity of continuing to follow an IvyVest approach to investing, and I continue to think that investing in a low-cost and diversified pool of ETFs and follow a set of rules for dynamic asset allocation is the best way for most people to invest. I may continue to keep the site up in some form or another as a free resource past this date, but will no longer manage saved model portfolios. I will have more to share in regards to this in the coming months.

If you came here for some light reading, you might be disappointed. This month we are going to discuss the “neutral rate of interest” and “secular stagnation.”

Why are we diving into a topic this wonky? Well understanding these things may help us to understand what we have often cited as one of the most significant feature of our financial world over the last thirty years: the persistent low interest rates that have persisted through numerous false-alarms about the "end of the bond bull market." These persistently low interest rates are a driver of much financial phenomena that we have experienced over the last couple of decades -- including the current high valuation of the stock market. So we should care a lot about why the exist and whether they are likely to continue.

Understanding interest rates may even help us to make some educated guesses about how interest rates might change in the future. This letter was motivated by and borrows heavily from a recent paper (1) by Lukasz Rachel of the Bank of England and Lawrence Summers of Harvard. Henceforth, we will refer to this paper as the RS paper. This era of low interest rates deserves much of the credit for the boom in asset prices we have seen over nearly the last forty years.

We will start with some definitions. The neutral rate of interest, also called the natural rate of interest, is the real (inflation adjusted) interest rate that allows for full employment and a desired low level of inflation. If actual interest rates (inflation adjusted) are higher than the neutral rate, they will tend to slow economic activity (and reduce inflation); if they are lower than the natural rate, they will tend to accelerate economic activity (and increase inflation). Secular stagnation is a state where the neutral rate of interest is very low, perhaps even negative. In this condition central banks may not be able to stimulate the economy by cutting rates. Current rates on government bonds are negative in many parts of the developed world, and further cuts may be practically impossible (after all, people can always take cash home and store in a mattress). So, some economists say we are now experiencing secular stagnation. Determining what the neutral rate is at any time is not easy. Central bankers spend a lot of time and effort trying to understand it, but they would probably be the first to admit that it can’t be determined exactly.

There is another way of thinking about neutral rates, and RS use it in their paper. The neutral rate can be considered as the rate where saving propensities equal investment propensities. Notice that we said savings propensities and investment propensities. Many of you probably realize that savings and investment always exactly balance according to the usual way of tracking a national economy. However, this is just a manner of bookkeeping. It doesn’t mean that productive investment exactly balances saving. For instance, a dollar deposited in a bank is savings for the depositor. If the bank lends it to someone who is going to invest it in a business, then savings equals investment. However, the bank may add the dollar to its excess reserves. In that case, the dollar becomes an investment on the bank’s books; total saving still equals total investment, but the investment is not productive. So, from this point of view, the neutral rate is the rate that brings savings into balance with real (productive) investments.

Regardless of what one thinks about the current level of neutral rates, actual market-based interest rates have been declining for a long time. Figures 1 to 3 show the real (inflation adjusted*) interest rate on 10-year government bonds in the U.S., Germany and Japan. In Germany and Japan, they are currently negative, and in the U.S., they are approaching zero. Although there have been fluctuations, the general picture is of a steady decline for 30 or more years.

image1.png Figure 1: Real interest rate on U.S. 10-year treasury bonds. The real rate is computed by subtracting the most recent annual inflation rate from the bond interest. Data is from the Federal Reserve Economic Data (FRED)

image2.png Figure 2: Real interest rate on 10-year German government bonds. Data is from FRED.

Figure 3: Real interest rate on 10-year Japanese government bonds. Data is from FRED.

Some of you may be thinking that interest rates are determined by the central banks (the Federal Reserve system in the case of the U.S.). That is only partially true. In the U.S., the Federal Reserve System (the Fed) determines very short-term overnight rates (the Fed rate) and it can try to influence longer term rates by buying or selling longer term bonds. However, its powers are often over emphasized. If it tries to set rates too high, it will depress the economy and create a recession or worst. If it creates rates that are too low, it creates inflation, which will eventually lead to higher rates and perhaps a severe slowdown. The central banks can’t deviate too much or too long from the neutral rate.

Low interest rates, therefore, cannot just be a function of central banks "printing money", for if that were the case it would simply create the larger mystery of why we have not seen any inflation yet. The persistence of low rates without inflation requires an explanation based on economic fundamentals. For investors, it’s important to know what factors are controlling the neutral rate so they can make educated guesses about what may happen to it in the future.

This topic has been the subject of much discussion. We discussed it in one of our newsletters some time ago. Now, RS have published results based on economic modeling. They admit that the modeling is an oversimplification of reality, but they are experienced economists and their conclusions are worth some thought. First, they had to make an estimate of what the neutral rate is now. There are several procedures in the literature for doing this, but we are not going to discuss them in detail here. The essence of the procedures is that an output gap is determined for the economy using historical trend data. The output gap is the difference between the gross domestic product (GDP) at any time and the maximum potential GDP consistent with stable inflation. By looking at the way that the output gap has varied over time, and by considering actual interest rates over time, inflation, and other factors, the models produce an estimate of the neutral rate of interest. Trade surpluses or deficits complicate these estimates. RS avoided this problem by considering the advanced economies, as defined by the International Monetary Fund (IMF), as a single entity. They note that the advanced economies, when considered as a group, have had very small current account surpluses or deficits. They also note that capital flows rather easily between the advanced economies, so it makes sense to consider an average neutral rate for the entire group. Using a procedure from literature, Summers and Rachel estimate that the current real neutral rate of interest for the advanced economies is about 0.5%. Further, they estimate that it has declined by about 4 percent points since early 1970s.

Figure 4, adapted from their paper, shows the decline in the estimated neutral rate. image4.png Figure 4: The real neutral rate for the advanced economies as determined by Rachel and Summers.

So, the question is, why has the neutral rate declined so much? RS answered this question by running two different economic models (general equilibrium models). We won’t discuss the models here, but both are well-known models from the economics literature. Their conclusions are interesting and worthy of consideration. Per RS, Table 1 shows how various factors have affected changes in the neutral rate since 1971. The general picture is that government actions have been pushing the neutral rate up, but private actions have been pulling it down. However, the entries on this table require much more discussion.

Table 1: Factors causing changes in the real neutral rate since 1971, per Rachel and Summers. The data was extracted from a graph, and the data points may not be highly accurate.

Line Number Factor Sector Percent change in neutral rate
1 Old age health care Government +1.1
2 Social security Government +1.2
3 Precautionary savings: Higher supply of assets Government +0.5
4 Government debt Government +.9
5 Total of government factors Government +3.6
6 Declining productivity Private -1.6
7 Lower population growth Private -0.9
8 Growing inequality Private -0.6
9 Longer retirement Private -1.1
10 Interactions between factors Private -1.2
11 Total of private factors Private -5.6
12 Total of all factors Government + private -1.7
13 Calculated change since 1971 -3.2
14 Unexplained change -1.5

Let’s discuss the government actions first. The first two lines in the chart deal with government payments to elderly and mostly retired people. Elderly people constitute a growing share of the population, and they are more likely to consume their income and less likely to save it. In economics speak, they have a high “marginal propensity to consume” (MPC). To the extent that they collectively receive a higher portion of society’s income, collective savings will decrease, and interest rates will rise. Government payments to the elderly also reduce savings in another way: Knowing that they will receive government assistance when they are older, younger people will tend to consume more and save less. Summers and Rachel estimate the growth of the elderly population and total government payments to the elderly has raised the neutral rate by about 2.3 percentage points (the sum of the first two lines in the Table). The next two lines can be a little confusing. The fact that increasing government debt increases interest rates seems fairly obvious: when the government borrows, it competes with private borrowers and drives up interest rates. Some of you may realize that this contradicts the doctrine of Ricardian equivalence, but we have always been very dubious about Ricardian equivalence, and the RS analysis indicates that it doesn’t hold. We admit that the distinction between lines 3 and 4 in the table isn’t totally clear to us, but we interpret them to mean that the net effect of increased government borrowing has been to raise the neutral rate by about 1.4 percentage points, and the total effect of government actions has been to raise the rate by about 3.2 percentage points. While government actions have been pushing the neutral rate up, aspects of the private economy have been pulling it down. Lines 6 to 11 deal with these effects. Lines 6, productivity growth, and line 7, population growth, both influence the growth rate of the economy. Both factors are lower now than in the past, and economic growth has slowed. When the economic growth rate is lower, there is less demand for investment capital, and this pulls the neutral rate down. Per the RS analysis, the net effect of these two factors is to pull rates down by about 2.5 percentage points. Inequality (line 8) also affects the balance between savings and investment. Inequality has been growing in the advanced economies. Wealthy people consume a smaller fraction of their income and save more.

So, as the income distribution shifts towards the wealthy, total savings in the economy increase, and interest rates tend to decrease. We were actually surprised that this was a relatively small effect (0.6 percentage points) in the RS analysis. We had expected that it would be a bigger effect. Line 9 refers to the fact that life expectancies have increased, but the average number of working years has not, so workers need to save more for retirement. The RS analysis indicates that this is pulling interest rates down by about 1.1 percentage points. Interactions between these factors (line 10) pulls the rate down by an additional 1.7 percentage points. Altogether, combining public and private influences, the analysis using the general equilibrium models indicates that the neutral rate should have dropped by about 1.7 percentage points since 1971. Since they their analysis of the neutral rate indicated a larger drop (3.2 percentage points), they conclude that their general equilibrium models are missing effects that total about -1.5 percentage points. The analysis is obviously very approximate, and RS are quite open about this.

Our interest in this analysis is in understanding why interest rates have been so low for so long and what may happen to them in the future. We confess that we had been thinking that rates were abnormally low and that rising government debt would drive them back towards more “normal” levels. The RS analysis makes us rethink this issue. Given the factors that RS considered, how are they likely to change in the future?

In the near future, total government indebtedness is likely to continue rising, and government spending on the elderly will continue to rise as the fraction of elderly in the population increases. It seems that government actions are likely to continue to put upward pressure on interest rates. Looking at the factors pushing downward, some are likely to continue to do so. Population growth will continue to slow in the developed countries. Many of them may hit population peaks in the not too distant future. Also, longer retirements are almost certain to continue. However, it is possible that some of these factors may reverse. One possibility is that total factor productivity could rebound as a result of developments in artificial intelligence or other technology developments. This could increase economic growth rates and the demand for investment capital. It is also quite possible that political actions may cause a reversal in the growth of income inequality. In the U.S., all the democratic candidates for president have proposals to reduce inequality, and even some republicans are discussing it. If the growth of inequality slows or reverses, net savings may decrease and interest rates may increase. There is another factor which should be discussed. In the past, we have assumed that large savings in China were putting downward pressure on global interest rates.

By focusing on the developed economies as a single unit, the RS analysis doesn’t consider this. Since the combined current account of the developed countries has been close to balance during the period of the analysis, RS may be justified in ignoring effects in other countries. Nevertheless, China’s high savings and strongly positive current account** has given them funds to invest abroad and has probably had a negative effect on global interest rates. The savings rate in China remains very high, but the current account as been coming down in the last few years. A continued decline in their current account could remove some downward pressure on interest rates.

On balance, we think a continuation of low interest rates is quite likely in the near term, but a rebound could be triggered by a higher rate of productivity growth, by political change in the U.S., or possibly by lower Chinese saving. If higher rates were triggered by productivity growth, it would probably be neutral for equities. Higher growth rates would improve future equity earnings, but the future earnings would be evaluated with a higher discount rate. Political actions to reduce inequality would most likely be negative for equities in the short term. Some of the excess savings that is moving into the equity markets would be removed. However, in the longer term the effect could be positive because it might increase demand for consumer products. If higher rates come about because China cuts back on its international bond purchases, that would probably be negative for equities. However, there would also be changes in exchange rates that make predicting the outcome very difficult.

So, while interest rates may remain low for a while, we think the picture could change in a few years. In particular, the election of a Democratic president in the U.S. could start the movement to higher rates. In any case, we continue to believe that investors should maintain a diverse portfolio that includes foreign and domestic equities, bonds, and commodities.

In these figures, inflation adjustment has been accomplished by subtracting inflation over the past year from the current bond yield. It would be better to subtract the inflation that is anticipated for the future, but that is much harder to do. *A nation’s current account is the balance of trade in both goods and services. It also includes cash transfers.

Reference 1) Lukasz Rachel and Lawrence H. Summers, “On Falling Neutral Real Rates, Fiscal Policy, and the Risk of Secular Stagnation”, Brookings Papers on Economic Activity, BPEA Conference Drafts, March 7–8, 2019.

Get our next article delivered to your inbox.

Sign up below and be the first to know about our freshest data-driven thinking on the markets, and investing. We will send you no more than one email a week. This is free.

Ready to start putting this into action?

Take a free two-week trial to IvyVest premium -- our premium subscription service. You'll get access to our rules-based dynamic asset allocation model, tools that will show you exactly what you need to buy in your own discount brokerage account (and when to re-balance) to implement it for yourself, and an insightful monthly newsletter that will keep you on abreast of the most important things going on in the markets. There is no credit card required. Get Started Now!

By Alex Frey