Monetary policy and bond yields

Here’s the FT:

A steady decline in yields since the start of the second quarter accelerated sharply this month, which market participants attributed to a liquidation of short positions by hedge funds and other momentum-orientated traders whose bets had turned against them. . . .

“The market was banking on a dovish contingency at the Fed,” Tipp said, who would allow the economy to run hot, pushing up inflation and reducing the value of long-dated bonds. That narrative stalled last month, he said, when Fed officials opened the door to raising rates in 2023, earlier than previously expected.

Mark Lindbloom, who manages the Western Asset core plus bond fund echoed that view. “We do not believe the Fed today, or in the future will sacrifice its credibility” from taming inflation in the 1980s, he said.

It seems like some bond traders had assumed that AIT was being used by the Fed as an excuse to engage in a more dovish monetary policy. Then, some recent statements by Fed officials convinced the market that the Fed is serious about keeping inflation close to 2% on average. Inflation expectations fell and this dragged down long-term bond yields.

In my view, there are still two unexplained mysteries:

1. While there are cases where a contractionary policy announcement immediately reduces bond yields, in other cases the effect seems to be delayed. Why is that?

2. What explains the long run downward trend in real interest rates? It’s clearly not (easy) central bank policy, as we’ve seen that tighter money actually reduces interest rates. So what is it?

Given point #2, the Fed needs to fix its monetary policy. While the adoption of AIT is a step in the right direction, a further step is needed. I see two primary options. First, raise the inflation target to 3%. Realistically, that’s not going to happen. Second, acquire a more powerful tool kit (a subject on which I’ve written extensively). Given the gridlock in DC, that probably won’t happen either, even though each party would probably support the idea at a time when they were in power. A couple years ago, I suggested that the Fed was missing a golden opportunity to ask for more powerful tools when Trump was in power. Now it’s too late.


Tags:

 
 
 

60 Responses to “Monetary policy and bond yields”

  1. Gravatar of PG PG
    11. July 2021 at 10:42

    Two applications of Occam’s razor:

    1) Best guess: risk sentiment and expectations. If everyone has one foot out the door and tightening is faster than expected, yields down right way. If sentiment remains positive (e.g. in a mania) and the tightening was expected, then no one will buy bonds until fear hits. (See https://money.cnn.com/2000/03/21/economy/fomc/ for example)

    2) Isn’t it just demographics? More risk-averse savers over time.

  2. Gravatar of Effem Effem
    11. July 2021 at 10:45

    Personally, I think #2 is one of the biggest questions in economics and doesn’t get enough attention.

    I view the risk-free rate as a function of the supply/demand for capital. Demand for capital is a function of economic growth, which we know is demographics + productivity + inflation. It’s trended lower, but not in any remarkable way.

    The supply of capital on the other hand has exploded higher (see the wealth/gdp ratio). I believe this is a function of shorter recessions and pro-capital policies such as bailouts. Historically, recessions wiped out a chunk of the capital stock keeping the supply and demand for capital roughly in balance. In recent decades we’ve done everything possible to short-circuit any capital destruction. This has some benefits, but also has the side effect of allowing the size of the financial capital stock to grow ever-larger relative to the size of the underlying economy. Hence a steady downtrend in the risk-free rate.

    We appear to now have created a system where the financial capital stock is so large that it dominates monetary policy. When the expected returns on capital are extremely low, a flight to cash is all the more probable (a modest fall in expected returns makes cash a superior asset). The threat of this sort of “unwind” is sufficiently scary that we have no choice but to react ever more quickly to any fall in asset prices. Thus, the upward trend in wealth:gdp continues.

    Once you hit the zero-bound you can no longer lower the return on cash (unless you employ negative rates or raise the inflation target), which makes this system all the more unstable. There is a mountain of “wealth” with expected returns barely above cash ready to convert to cash at any sign of trouble.

  3. Gravatar of Lewis Johnson Lewis Johnson
    11. July 2021 at 11:20

    Falling real rates IS the most important question in markets. I wrote the below on this topic 5 years ago.

    https://www.capitalwealthadvisors.com/2016/05/end-blanche-dubois-market/

  4. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 13:32

    My first thought on question 1 is that the markets only sometimes think the Fed will lower rates in response to new disinflationary expectations. AS you’ve pointed out many times, the Fed is almost always behind the curve when it comes to easing policy.

    Also, occasionally, typically in panics, there are also liquidity problems in the Treasury markets.

    On question 2, while I think there are some real factors at play with the downward trend in real interest rates, they are overestimated. We shouldn’t underestimate central-bank induced falling inflation rates, particularly when many large central banks turn relatively hawkish around the same time, such as since the end of the Great Inflation.

    There is something very wrong about how economists view monetary policy in the international context, just as there are poblems with how it’s viewed in the domestic context, and the combination of problems has led to failing to appreciate just how low the nominal and real equilibrium rates often are, and hence how long it will take for economies to recover after slowdowns and recessions. Recoveries have gotten progressively longer, and termed “jobless” since the recovery after the 1990 recession.

    It’s easy to look at long-term trends and say they must be secular, until one remembers that all developed country central banks and many besides sought to lower inflation after the Great Inflation, and that they have 100% control over interest rates.

    Even with the demographic slowdowns in most of the relevant countries for the current discussion, there’s evidence that, in the US anyway, innovation actually picked up in the 90s and 2000s. In the late 90s, the growth in innovation clearly swamped the coming expected demographic slowdowns, and I suspect the same would be true today, though to a lesser extent, if we could achieve and maintain monetary equilibrium. US TFP growth increased during the 90s and 2000s.

    I think the economics profession will eventually accept the metrics I’m currently using to help identify periods of monetary disequilibrium. It should be clear, logically and empirically, that when stock prices, for example, are increasing more rapidly than earnings for a decade, then money was tight for that decade. It means short-term earnings are depressed by tight monetary policy, while prices reflect outer future years in which the economy healed due to gradual wage adjustment. And, if you accept this logic, then it is clear that real interest rates would be higher sooner, if NGDP growth were higher, sooner.

    By contrast, during the 70s, nominal stock prices were basically flat, indicating that the earnings yield was rising pretty much in lockstep with prices, that rising inflation expectations were being factored into earnings immediately. In real terms, stocks actually declined in value in the 70s.

    This all fits my model, which I remind anyone reading did predict much lower unemployment rates were possible during the last recovery than most economists thought possible.

  5. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 13:42

    Effem,

    You wrote:

    “When the expected returns on capital are extremely low, a flight to cash is all the more probable (a modest fall in expected returns makes cash a superior asset).”

    I think you’ve reversed the direction of causation. Causation goes from tight money that causes an expected increase in the future value of the dollar versus that of output and capital returns, and this crushes capital asset values in a very predictable way.

  6. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 14:03

    By the way, it’s become humorous to me now when various commentators, some of whom are well-appointed, talk about stock market “bubbles”, because P/E ratios are high, particularly versus some arbitrary baseline. Here’re Larry Summers and Ray Dalio expressing such concerns, along with their very misplaced concerns about high inflation risks:

    https://www.youtube.com/watch?v=Xa2TrSYCJI4

    It’s easy now to spot those who’ve never studied the relationship between stock prices and economic growth. If they had even a basic understanding, they’d realize that stock earnings are significantly more volatile than stock prices, and hence as economic growth picks up, earnings will rise disproportionately to overcome the increase in the discount rate. Look at this data, for example:

    https://www.macrotrends.net/1324/s-p-500-earnings-history

    The truth is literally the opposite of what people like this claim. Stock prices are set to rise as wages adjust and/or monetary policy loosens during an otherwise slow recovery. Slow recovery = low rates for longer.

    I don’t think even Robert Shiller understands this, and he’s supposed to be one of the foremost experts on the stock market among economists.

    The economics field is truly in the dark ages.

  7. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 14:19

    Perhaps I should be clearer on the difference between the 70s and recoveries during tight money periods. I explained the 70s situation, which was obviously a loose money period. Once inflation expectations set in, the current increases in inflation are deemed permanent, causing the discount rate and stock prices to rise in tandem.

    In contrast, when money is tight, and then loosens, for example, expectations for increased relatively volatile earnings growth in sooner years boost stock prices disproportionately vis-a-vis the discount rate.

  8. Gravatar of Effem Effem
    11. July 2021 at 14:27

    @Michael

    Tight money causes things like high P/E ratios and tight credit spreads? Seems the opposite to me. Policy, on average, has been neither tight nor loose since we’ve basically tracked ~2% inflation over the last few decades. And yet expected returns are compressed on every major asset class.

  9. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 14:55

    Effem,

    When markets perceive that monetary policy has suddenly tightened, stock prices fall, lowering P/E ratios, as earnings in the ratio are for the trailing 12 months. However, after the next earnings report, earnings will fall more than stock prices, which leads to higher P/E ratios than before policy tightened.

    Then, the recovery can take different shapes. If NGDP level targeting were suddenly, credibly adopted, earnings would rise faster than prices, until the P/E ratio settled in equilibrium with the expected NGDP growth rate. With a lack of such a rapid return to the prior trend NGDP growth rate, stock price appreciation would lead earnings growth, for reasons described above. Tight money depresses current and shorter-term economic growth, and hence earnings growth, but longer-term expacted growth is higher, because the economy is expected to recover as wages adjust, if for no other reason. That would lead to higher PE ratios during weak recoveries.

    It’s not the way most people think about this, but then most people are wrong. The logic and evidence are very clear.

  10. Gravatar of ssumner ssumner
    11. July 2021 at 14:56

    Michael, You said:

    “On question 2, while I think there are some real factors at play with the downward trend in real interest rates, they are overestimated. We shouldn’t underestimate central-bank induced falling inflation rates, particularly when many large central banks turn relatively hawkish around the same time, such as since the end of the Great Inflation.”

    I’m not sure what you are saying here. What does central bank policy have to do with long run trends in real interest rates?

  11. Gravatar of Matthias Matthias
    11. July 2021 at 15:08

    Michael, does your preferred metric correct for the different effects on the stock price of dividends vs buybacks?

    Economically they are the same, so you metric should correct for any differences.

    Btw, if you have lots of buybacks or dividends, stock prices can rise quicker than earnings without much trouble.

    Or more meaningfully, total stock market returns can be above earnings for a long time.

    Eg imagine a company with flat earnings that decreases its total market capitalisation via buybacks and perhaps leverage over time.

    Alternatively, think through a company with flat earnings that never pays a dividend nor buys back any stock and perhaps even raises more and more capital.

  12. Gravatar of dtoh dtoh
    11. July 2021 at 15:19

    Scott,

    Regarding the mysteries…

    1. This is obvious. Fed policy/action can cause a shift of the curve and/or movement along the curve. Unless you know if and by how much the curve has shifted, yields alone don’t tell you anything.

    2. Real rates may not be declining over time. We might just be over-measuring inflation so real rates are actually flat of increasing.

  13. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 15:56

    Scott,

    I apologize if my point wasn’t clear. I guess it was clear how it related to nominal rates, but not real rates. I reference this graph:

    https://fred.stlouisfed.org/graph/?g=Fl1U

    Notice how, except for the 1990 recession and long recovery, real rates were stable during the Great Moderation, roughly 1983 to 1999. Negative real rates became common after the 2001 recession, consistent with tight monetary poliy.

  14. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 16:06

    Mattias,

    You asked:

    “Michael, does your preferred metric correct for the different effects on the stock price of dividends vs buybacks?”

    I haven’t researched this question specifically, but off the top of my head, ceteris paribus, stock buybacks would be preferred during tight money recoveries, as companies, along with the rest of the market, expect higher future earnings growth rates, and hence further stock price appreciation. Why increase dividends when earnings are depressed? However, differential tax treatment means ceteris paribus doesn’t apply here.

    In the long run, the mean S&P 500 earnings yield is very close to the mean NGDP growth rate, so companies behave in ways to reach that equilibrium in the long run.

  15. Gravatar of Michael Sandifer Michael Sandifer
    11. July 2021 at 19:06

    “By contrast, during the 70s, nominal stock prices were basically flat, indicating that the earnings yield was rising pretty much in lockstep with prices,…”

    I need to correct that sentence from above. Instead of “prices”, I meant “future earnings”.

    I point out that all I’m doing here is relating short-run and long-run stock earnings growth to SRAS/SRAD and LRAS/LRAD. There’s nothing new in that sense to basic macro theory. What’s new, in a sense, is that the earnings yield is the same as the expected NGDP growth rate, in monetary equilibrium.

    It means we can level target NGDP using the S&P 500 P/E ratio. A 5% NGDP target would require a 20 P/E.

    And if you think about it, even when there are stock-specific influences on stock prices, such as increased taxes on stock dividends or capital gains, it should just lower the price path of the S&P 500 on a one-time basis, but prices will continue to rise at the expected NGDP growth rate, if in monetary equilibrium. Basically, you can level target the S&P 500 price growth rate.

    Empirically, this seems to be supported, as P/E ratios naturally return to their equilibrium levels in the longer run. I think they would do so in the short-run under NGDP level targeting.

  16. Gravatar of Lizard Man Lizard Man
    11. July 2021 at 19:26

    Isn’t the vast majority of borrowing used to construct physical assets? Has the construction of physical assets trended downwards over the past few decades in the developed and near developed countries?

    I would expect that the majority of the world’s new physical assets are being built in developing and middle income countries, where much less money is needed to construct that infrastructure. So if you aren’t busy building lots of roads, houses, railways, factories, etc., why do you need to borrow money?

  17. Gravatar of James Alexander James Alexander
    11. July 2021 at 22:43

    Dtoh +1
    Over-estimating inflation. The amount of time consumers spend on stuff that is essentially free (to them) must have increased substantially: ie stuff on the internet. How is free stuff captured in PCE Price indexes (or CPI)? Broadband or mobile data pricing is a very narrow way of capturing the immense breadth of what the internet provides.

  18. Gravatar of James Alexander James Alexander
    11. July 2021 at 23:05

    Well, using google (for free) I have (spent time and) found some pieces that indicate there is some discussion but no consensus and certainly little or no impact on official data. But, personally, I have gained some consumer satisfaction, for free, from the activity.

    https://www.philadelphiafed.org/-/media/frbp/assets/working-papers/2016/wp16-24.pdf
    https://www.sipotra.it/wp-content/uploads/2019/03/Measuring-consumer-inflation-in-a-digital-economy.pdf
    https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3170839

    NB the last one is behind a paywall, looks interesting, but isn’t free, so I didn’t “purchase” it.

  19. Gravatar of Todd Ramsey Todd Ramsey
    12. July 2021 at 07:43

    Scott

    I’m not sure I understand:

    Do you think “the long run downward trend in real interest rates” is BAD (something the Fed needs to “fix”)?

    If so, why?

  20. Gravatar of Todd Ramsey Todd Ramsey
    12. July 2021 at 07:50

    Dear Strict EMH adherents:

    Did you see the Coval-Hirshleifer-Shumway study of individual investors linked by Tyler Cowen?

    What are the holes in the study that I am missing? Thanks in advance.

    https://academic.oup.com/raps/advance-article-abstract/doi/10.1093/rapstu/raab017/6311677?redirectedFrom=fulltext

  21. Gravatar of Michael Sandifer Michael Sandifer
    12. July 2021 at 08:18

    Todd Ramsey,

    All of the EMH adherents I’ve seen comment here, including Scott, acknowledge that there must be some arbitrage opportunities available to provide incentives to make markets efficient. “Efficient” in this case, is relative. Very few people can be markets consistently, on average, over long stretches of time, so markets make better predictions than the vast majority of investors.

  22. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 08:25

    Interest rates reflect the supply of and demand for loan funds, not the supply and demand for money. Essentially the deceleration in yields is a function of velocity. Velocity has fallen in lock step with the idling of savings in our payment’s system. As Vt falls, so does AD, and thus N-gDp. Japan’s economy is prima facie evidence.

    “The Savings Glut Is Really A Dearth of Investment”
    “https://www.bloomberg.com/news/articles/2020-12-11/the-savings-glut-is-really-a-dearth-of-investment

    All this was predicted in 1961. “The growth of commercial bank-held time “savings” deposits shrinks aggregate demand and therefore produces adverse effects on gDp”…

    The smartest guys in the room just don’t get it: Take Dr. George Selgin (advisor to Congress, Ph.D., Economics, New York University,1986): July 20, 2017

    “This is nonsense, Spencer. It amounts to saying that there is no such things as ‘financial intermediation,’ for what you claim never happens is precisely what that expression refers to.”

    It is an incontrovertible fact. Banks don’t loan out the savings of the public. Banks pay for their earning assets with new money.

  23. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 08:26

    There are six factors at play here that explain the 2nd quarter decline in longer term interest rates.

    The U.S. Treasury sold a large amount of bonds in March above and beyond what was already required to cover the structural budget deficit-roughly $342 billion by my estimate. This explains the peak in the 10-year bond yield in mid-March. The Treasury had a lot of additional needed borrowing out of the way before the 2nd quarter.

    The U.S. Treasury exited the 1st quarter with $1.1 billion in its bank account at the Federal Reserve (Treasury General Account). On June 30th it was $852 billion. The Treasury thus had $250 billion of additional funds from drawing down this account.

    U.S. Commercial banks continued their heavy buying in the 2nd quarter to the tune of $301 billion.

    Foreigners started buying again in April (+$42 billion) after being net sellers in March (-$71 billion). Foreigners have not been net buyers of Treasuries since July 2020 so this is an interesting development. May and June data is not out yet, but I expect that we are going to see further upticks in foreign buying.

    The U.S. Government still has a lot of spending to do to fund the commitments made in recent spending bills and this means that there is still a lot of borrowing to do-$2.3 billion at the end of May by my estimates.

    The Federal Reserve continues to buy $80 billion of U.S. Treasuries per month.

    I estimate that the Treasury only had to sell $141 billion in net new Treasuries during the 2nd Quarter. When you run the numbers, you can see that there were simply a lot more buyers than sellers. This rosy scenario ends in the 3rd quarter. A test is coming.

  24. Gravatar of ssumner ssumner
    12. July 2021 at 08:29

    dtoh, 1. That doesn’t explain why the response is delayed, if it is delayed.

    2. Any mismeasurement of inflation is far too minor to explain the sharp fall in real yields. And even if you are right, we’d then just have a mystery of why nominal yields have fallen more than NGDP growth. So the mystery does not go away.

    Michael, Real rates have been falling for 40 years—that’s clearly not monetary policy.

    James, See my reply to dtoh.

    Lizard, As I recall, the best evidence is that the investment schedule shifted left and the saving schedule shifted right, resulting in lower rates but not much change in S&I as a share of GDP.

    Todd, No, I don’t think it’s bad, and the Fed couldn’t “fix” it even if it wanted to.

  25. Gravatar of Michael Sandifer Michael Sandifer
    12. July 2021 at 08:39

    Scott,

    Did real rates fall in the US during the Great Moderation, 1990 recession aside?

  26. Gravatar of James Alexander James Alexander
    12. July 2021 at 11:08

    It’s possible that NGDP growth has been so below trend because monetary policy has remained very tight, thus explaining low nominal rates. At least true RGDP will have been higher than estimated.

    I don’t see why in the face of such huge technological change RGDP measurement (essentially finding the correct amount for the deflator) would have kept pace. The market capitalisation of Alphabet, FaceBook, Apple and Microsoft is unprecedented, and better evidence of the changing economy.

  27. Gravatar of James Alexander James Alexander
    12. July 2021 at 11:09

    And Amazon, of course.

  28. Gravatar of James Alexander James Alexander
    12. July 2021 at 11:10

    And Amazon, of course!

  29. Gravatar of Jeff Jeff
    12. July 2021 at 11:36

    Mis-measuring inflation can’t account for the drop in real rates. The keepers of the major price indices (CPI, GDP deflator, etc.) have tried, over time, to make more and better adjustments for hedonics, quality, substitution, etc. The usual effect of these adjustments is to lower estimated inflation, mostly in more recent periods. The overall effect from such adjustments is to induce a downward trend in inflation, which translates into rising rather than falling real rates. To get a substantial effect going the other way, you’d have to posit that inflation 50 years ago was substantially underestimated compared to today. All the evidence I’ve ever seen says the opposite.

    There’s a lot to the demographic argument. Between China and the US, huge numbers of people are trying to save for impending retirement. All those savings seeking good returns push those same returns down.

    But it seems to also be true that physical capital, the kind that can be purchased, is less important to a modern economy than human capital and expertise. You can buy a tech company, but you can’t exploit it the way you can, say, an improved iron smelter. If you try to, the tech guys just quit and go to work somewhere else. So, productivity continues to grow, but the drivers of that productivity can’t actually be purchased by investors.

    Both of these changes (demographic and the nature of capital) take place very slowly and over decades. Much like the declining real rates do.

  30. Gravatar of Brian Donohue Brian Donohue
    12. July 2021 at 11:42

    on #2, I used to think the human propensity to discount the future means a positive real interest rate was a law of nature, but there is a countervailing interest among old people with money to preserve, to the extent they can, the purchasing power of accumulated assets. When Boomers were young, the first factor dominated, but now, and at least for the next decade, the second factor dominates. Presto, negative real rates.

  31. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 12:01

    People act like the U.S. is somehow different. In the circular flow of income, in a closed circuit, interest rates are determined by the supply and demand for loan funds. Supply is fed by savings.

    But in the real economy, the closed circuit is circumvented by the injection of new credit by the FED. Thus supply is increased. But at the same time demand is reduced by the decline in velocity (by the impoundment and ensconcing of monetary savings).

    COMMERCIAL BANKS AND FINANCIAL INTERMEDIARIES: FALLACIES AND POLICY IMPLICATIONS-A COMMENT LELAND J. PRITCHARD

  32. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 12:04

    re: “I see two primary options. First, raise the inflation target to 3%.”

    FAIT is already set to kill the exchange rate of the U.S. $.

    Remember the stagflationists

    Link: “Rethink 2%” http://bit.ly/2s67De9

    What do you think is different this time around?

  33. Gravatar of Effem Effem
    12. July 2021 at 12:12

    Another way to look at this is simply that the “velocity of wealth” is very low (low propensity to spend among the wealthy is usually pointed to). Massive amounts of wealth have had to be created (via reductions in the discount rate) in order to keep economic activity on target.

    Which begs the following question: should the “velocity of wealth” increase would the Fed have the political leeway to essentially destroy wealth to keep inflation in-check?

  34. Gravatar of ssumner ssumner
    12. July 2021 at 12:39

    James, Yes, NGDP growth has slowed, but that’s not the issue. Nominal interest rates are surprisingly low even relative to (falling) NGDP growth rates. So I don’t think mismeasurement of inflation is the issue. Any mismeasurements affect estimates of real rates and real GDP growth equally.

    Jeff, Good comment.

    Brian, I agree.

    Effem, I think you mean “propensity to consume”. And the Fed can and will keep inflation in check, after a brief period of high inflation this year and perhaps part of next.

  35. Gravatar of Kester Pembroke Kester Pembroke
    12. July 2021 at 12:41

    http://bilbo.economicoutlook.net/blog/?p=47883

    Investors lose out following New Keynesian advice.

  36. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 13:02

    re: “The market capitalisation of Alphabet, FaceBook, Apple and Microsoft is unprecedented”

    That’s why Biden “signs order to crack down on Big Tech, boost competition ‘across the board’”

    “Despite that fact that job openings are at all-time high, at the same time, initial jobless claims are running at a level that is 75% above/higher than the pre-pandemic level. This begs the question: why is the US laying off people at a rate that is three quarters greater than before COVID-19, despite that fact that the economy has reopened?”

  37. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 13:44

    “What we have is peak consumption of things that aren’t made in the USA.”

  38. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 14:10

    There is an extraordinary empirical stability and regularity to such magnitudes as income velocity that cannot but impress anyone who works extensively with monetary data” (Friedman, 1956, p. 21).

    Friedman didn’t know what he was talking about. What happened since the mid 1960’s was that the federal funds rate became the sole criterion in the formulation and execution of open market policy. As a consequence, once accelerating inflation started in the mid 1960’s, and market rates began to rise, reflecting an inflation premium, the system purchased government securities in increasing quantities in an attempt to hold interest rates at their prevailing levels.

    These purchases were reflected in the rapid expansion of Reserve Bank credit, which increased from a level of $39.9 billion at the end of 1964 to $91.6 billion at the end of July 1974.

    By using the wrong criteria (interest rates rather than member bank reserves) in formulating monetary policy, the Federal Reserve became the engine of inflation.

    Using the transaction’s velocity of money in American Yale Professor Irving Fisher’s truistic equation of exchange, during the decade ending in 1964 aggregate monetary demand increased at an annual compounded rate of about 6 percent. In the subsequent 9 years, the increase was more than 13 percent, and in 1972-73 nearly 30 percent.

    I.e., there’s a monumental difference between income velocity and the transactions’ velocity of money. 1978 is prima facie evidence.

    There were 27 price forecasts by individuals and 9 by econometric models for the year 1978 (Business Week). The lowest (Gary Schilling, White Weld), the highest, (Freund, NY, Stock Exch) and (Sprinkel, Harris Trust and Sav.).

    The range CPI, 4.9 – 6.5 percent. For the Econometric models, low (Wharton, U. of Penn) 5.7%; high, 6.6% U. of Ga.). For 1978 inflation based upon the CPI figure was 9.018% [and Leland Prichard, in his Money and Banking class, predicted 9%].

  39. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 14:30

    In other words, the demand for money (transactions’ velocity) will determine the course of interest rates.

    Joshua Hall “The reality is, the Fed and commercial banks are the ones doing all the buying. On a net basis, investors are not buying Treasury bonds at all.”

  40. Gravatar of Spencer Hall Spencer Hall
    12. July 2021 at 15:17

    Milton Friedman, for instance, explains that “Fisher, in his original version, used T to refer to all transactions – purchases of final goods and services…, intermediate transactions…, and capital transactions (the purchase of a house or a share of stock).

    In current usage, the item has come to be interpreted as referring to purchases of final goods and services only, and the notation has been changed accordingly, T being replaced by y, as corresponding to real income” (Friedman, 1990, p. 38).

    Werner “But the empirical record of financial deregulation was to increase the volume of transactions, suggesting a higher speed of transactions.”

    But both short-term and long-term monetary flows, volume times transaction’s velocity (AD), fell below zero during the July 1990 –Mar 1991 recession. As the time series demonstrates, financial transactions are not random.
    Monetary Flows { M*Vt }: Monetary Flows { M*Vt } 1921-1996

  41. Gravatar of James Alexander James Alexander
    12. July 2021 at 21:22

    This idea of real returns falling due to demographics and crowded investments is weird. Real returns have never been higher. Real returns on money, cash and bonds, are lower, but on real assets (stocks and real estate) it’s high. Many think it’s a bubble, even (irony alert).

  42. Gravatar of James Alexander James Alexander
    12. July 2021 at 21:28

    Jeff: “ So, productivity continues to grow, but the drivers of that productivity can’t actually be purchased by investors.”
    Err…a NASDAQ ETF?

  43. Gravatar of Effem Effem
    13. July 2021 at 05:36

    @Scott
    “And the Fed can and will keep inflation in check, after a brief period of high inflation this year and perhaps part of next.”

    But will the Fed make up for any overshoot (as specified by their AIT framework)?

  44. Gravatar of ssumner ssumner
    13. July 2021 at 08:26

    James, Many people define “bubble” as a period of low expected real returns.

    Effem, That remains to be seen.

  45. Gravatar of James Alexander James Alexander
    13. July 2021 at 09:33

    Scott, pre- or post-bubble?

    Though I am surprised that you think bubble-predictors have that much sophistication. I thought they expected us to return to the Stone Age, after all “what goes up must go down”. Although the jeremiahs of the Stone Age often thought stone tool prices a bit racy.

  46. Gravatar of James Alexander James Alexander
    13. July 2021 at 09:37

    I can’t find a convincing chart of LR (100 years, say) Real Interest Rates. Looks to be Barry above 0% in the LR.

  47. Gravatar of James Alexander James Alexander
    13. July 2021 at 09:39

    Looks to be barely above 0% in the LR, ie well within any realistic Margin of Error.

  48. Gravatar of Effem Effem
    13. July 2021 at 09:44

    @Scott

    “Effem, That remains to be seen.”

    This is the problem. The Fed goes through a big exercise to redefine their framework and they come up with AIT. Then we get an inflation overshoot and the Fed repeatedly tells us its not a problem and there has been no change to their framework. And now all of a sudden we are in “remains to be seen” mode.

    Instead of spending all of their time debating what minor tweaks we should make to some theoretical framework or target, maybe economists should figure out how to better insulate the Fed from the whims of politics, or wanting to be liked, or whatever it is that would cause the Fed to almost immediately abandon a new framework.

    Any future frameworks will need a giant asterisk: *may be abandoned immediately if slightly inconvenient.”

  49. Gravatar of Ricardo Ricardo
    13. July 2021 at 15:56

    Economic Perspectives, No. 1, March 2021
    The Global Saving Glut and the Fall in U.S. Real Interest Rates: A 15-Year Retrospective

    https://www.chicagofed.org/publications/economic-perspectives/2021/1

    “return on capital appears … stable” but “safe and riskless rates fell sharply”

    One issue to be aware of is that when you say “real rates” what you mean is rates on US Treasurys.

    Rates on Treasurys are lower because of the shift of relative preference of savings (coming from savings-glut countries) into Treasurys vs. other assets.

    While yields on capital overall are stable, apparently… which was a bit of a surprise to me.

  50. Gravatar of dtoh dtoh
    13. July 2021 at 19:16

    @Scott,

    1. If you believe in EMH, there can be no delay in reaction. You just can’t see the reaction, because it occurs in the form of a shift in the curve. If you’re looking only at yields you won’t see it.

    Regarding my second point… Look at this way. If you are an investor, you’re trading current consumption for future consumption. If official inflation measurements have increasingly underestimated improvements in quality, then from a theoretical point of view, it can explain why investors are accepting a lower “real” (i.e. nominal less official inflation) rate. You can argue about the magnitude or existence of such an effect, but I don’t think you can deny its theoretical plausibility. And…. I don’t think you or anyone has come up with a better alternative explanation.

    Also I don’t see why there is any contradiction between that and a widening of spreads between nominal yields and nominal GDP growth. It may be a mystery, but it has no bearing on whether or not mis-measurement of inflation is the cause of the decline in “real” yields.

  51. Gravatar of dtoh dtoh
    13. July 2021 at 19:17

    @Michael Sandifer

    “It should be clear, logically and empirically, that when stock prices, for example, are increasing more rapidly than earnings for a decade, then money was tight for that decade.”

    Agree. I think this is totally obvious… but apparently not to most people. I think I may have commented on this a few years back.

    “It means we can level target NGDP using the S&P 500 P/E ratio.”

    Agree. What do you see as the advantage over simply targeting nominal GDP?

  52. Gravatar of Effem Effem
    14. July 2021 at 04:41

    @Ricardo

    Interesting paper. So if returns on private capital don’t fall with the risk-free rate then that must mean the entire burden of falling productivity hits labor?

  53. Gravatar of ssumner ssumner
    14. July 2021 at 09:51

    James. One definition of bubble prices is asset prices so high that the expected return going forward is unreasonably low.

    Effem, We will never reach a point of zero uncertainty about the future.

    Ricardo, In my view, expected yields on capital going forward are falling. Ex ante, real yields have recently been high, but I don’t expect that to persist. Nonetheless, I agree about the growing preference for safe assets.

    dtoh, You missed the point. If this was the explanation, then the relationship between NGDP and nominal rates should be stable.

    Of course I agree on the EMH.

  54. Gravatar of Michael Sandifer Michael Sandifer
    14. July 2021 at 11:09

    dtoh,

    The purpose of targeting the S&P 500 is to just give the Fed a forward NGDP growth indicator. This has the benefit of avoiding the step of creating an NGDP futures market.

    Of course, Unfortunately, this idea may cause even more political problems than creating the futures market, as I can imagine critics claiming the Fed is just trying to pump up Wall Street, as dumb as such criticism would be.

  55. Gravatar of Michael Sandifer Michael Sandifer
    14. July 2021 at 11:14

    Scott,

    You seem to dismiss claims that the fall in real rates is primarily due to monetary policy pretty readily for someone who, while solid on the theory he has, does not have a complete macro theory.

    Real rates were pretty flat during the Great Moderation, and fell sharply during periods of tight money thereafter.

    How do you explain the fall in long-term real and nominal rates after the initial Great Recession recovery began? Did the world suddenly wake up to the reality of slower US demographics?

  56. Gravatar of dtoh dtoh
    14. July 2021 at 15:29

    Scott,
    Why? There is no reason why constant real rates require that the spread between nominal rates and nominal growth must also be constant. Or for that matter, that the spread between real rates and real growth must be constant.

    The fact that price stability was maintained pre-Covid with rates of unemployment lower than most economists or policy makers thought possible suggests that either the Wicksellian Rate has changed and/or that policy was too tight. (Again this may or may not have been the case, but I don’t think it’s inconsistent with theory.)

  57. Gravatar of Michael Sandifer Michael Sandifer
    14. July 2021 at 17:18

    dtoh,

    Yes, there were those of us here all along saying there was still an output gap after 2015 and ’16, and that unemployment would continue to fall. But, true to form, the stupid Fed ignores the steepening SRAS curve as RGDP approaches sustainable capacity. They see inflation rise above their arbitrary 2% target and kill the party just as it’s getting started. Who needs rising real wages?

    Even my model predicts that wage adjustment should should lead to a full recovery about 6 years after the Great Recession, but only if there isn’t additional tightening along the way, and full recovery only refers to recovery from that particular shock. It can take longer, if the economy is alr already out of equilibrium before the recession begins, which was the case in this example.

  58. Gravatar of Jeff Jeff
    15. July 2021 at 11:37

    @James
    A NSADAQ ETF does not give you a claim on the money paid to employees. When it’s employees that drive productivity rather than physical capital owned by the firm, investors aren’t very important, and their returns are limited. There are exceptions, such as firm-owned intellectual property, but even there you find that firms with a lot of valuable intellectual property end up paying their employees more than firms that don’t have valuable IP. (The same used to be true of firms in capital-heavy industries, but I suspect the effect is more important in tech firms.) Some of the returns you might expect to go to investors are diverted to employee and manager compensation.

  59. Gravatar of Todd Ramsey Todd Ramsey
    16. July 2021 at 06:12

    Michael Sandifer-

    It seems to me Scott’s comment below implies that his view of the EMH says that individual investors cannot beat the market:

    “And I was told that the smart guys who ran the Harvard endowment could consistently outperform the market . . .

    . . . until they couldn’t. “

    I’m probably just not understanding. Can you explain your comment more completely?

  60. Gravatar of Michael Sandifer Michael Sandifer
    16. July 2021 at 14:41

    Todd Ramsey,

    Sure. The version of the EMH that seems most widely accepted in my bubble is the one that says markets are not absolutely efficient, in that there are market shocks, so market predictions aren’t always correct. It says that markets are instead relatively efficient. That is, they are efficient relative to the vast majority of market participants. It is very difficult, but not impossible for market partipants to consistently earn above-market returns for extended periods of time, on average.

    If markets were absolutely relatively efficient, what incentive would there be for, particularly, the smartest investors to invest in individual stocks? And without their participation, markets would be inefficient, creating opportunities for above-market gains. If all investors were to invest in the S&P 500, there would be plenty of stocks outside of that index that would be undervalued. Milton Friedman summed it up thusly:

    “If the market were 100% efficient, nobody could make any money making it efficient, and then it wouldn’t be efficient again.”

    Scott has made many comments in the past that indicate he doesn’t see the stock market, for example, as being absolutely efficient, nor absolutely relatively efficient, but mostly relatively efficient. T

    I see beating the market as similar to trying to have a career in music, comedy, or as an athelete. Very tiny percentages of people are able to earn even average livings in those pursuits, yet millions try. Those millions who try and fail create competition for those who succeed, which makes the successful better at their professions than they would otherwise be. The same is true of professional investors, such that even under-performing investors help make markets more efficient. Every bad investment decision, in complete markets, creates an opportunity for a very good one.

Leave a Reply