Why have real interest rates fallen so much?

From Lukasz Rachel and Thomas D Smith at the Bank of England (pdf):

Long-term real interest rates across the world have fallen by about 450 basis points over the last 30 years. The co-movement in rates across both advanced and emerging economies suggests a common driver: the global neutral real rate may have fallen. In this paper we attempt to identify which secular trends could have driven such a fall. Although there is huge uncertainty, under plausible assumptions we think we can account for around 400 basis points of the 450 basis points fall. Our quantitative analysis highlights slowing global growth as one force that may have pushed down on real rates recently, but shifts in saving and investment preferences appear more important in explaining the long-term decline. We think the global saving schedule has shifted out in recent decades due to demographic forces, higher inequality and to a lesser extent the glut of precautionary saving by emerging markets. Meanwhile, desired levels of investment have fallen as a result of the falling relative price of capital, lower public investment, and due to an increase in the spread between risk-free and actual interest rates. Moreover, most of these forces look set to persist and some may even build further. This suggests that the global neutral rate may remain low and perhaps settle at (or slightly below) 1% in the medium to long run. If true, this will have widespread implications for policymakers — not least in how to manage the business cycle if monetary policy is frequently constrained by the zero lower bound.

I believe this paper is being presented at Brookings, sometime early in March…


High profits require limiting the building of productive capital assets.

If labor is not paid to build capital assets the the economy lives off the "savigs" in labor stored in existing capital assets which are being consumed but not replaced.

The society of engineers are now placing the deficiency in public productive capital assets at 4 trillion. That's the under investment in road, rail, air, water, sewer, power, telco, ...

Burning capital is literally what the fossil fuel industry does, based on the need to save labor because labor is extremely scarce to grow the economy. To not burn capital, tens of millions of workers would need to be found who can earn middle class wages in manufacturing and construction, but the world is faced with desperate labor shortages, and can't afford to build wind and solar harvesting and storage capital assets. Where would the workers be found with the worker to population ratio so high that building more productive assets would require child labor and forcing 75 years old to work construction! ;-)

High profits require limiting the building of productive capital assets.

CapEx is captured on the statement of cash flows and balance sheet, not the income statement. Why do you consider yourself qualified to comment when you do not even understand the very basics of accounting?

If labor is not paid to build capital assets the the economy lives off the “savigs” in labor stored in existing capital assets which are being consumed but not replaced.

Labor theory of value, really?

Burning capital is literally what the fossil fuel industry does

What? The fossil fuel industry exists because its form of capital (oil) can be transformed into other useful forms of capital. If oil couldn't be transformed into something actually desired, it would not be capital. That is exactly why it wasn't considered capital for nearly all of human history.

Comments for this post are closed

Comments for this post are closed

"Meanwhile, desired levels of investment have fallen as a result of the falling relative price of capital"

This sentence makes no sense.

'Desired levels of investment have fallen' because of the declining returns on investment in all industrial sectors, which Robert Brenner explored almost ten years ago:


“Meanwhile, desired levels of investment have fallen as a result of the falling relative price of capital”

Agree that it is confusing and perhaps poorly worded. Here is an attempted re-phrasing: "Implicit in the "price of capital" is the return on capital. This implies that when the relative price is high (as compared to say labor) is high, returns are expected to be high. If the relative price of capital is falling, this would indicate to would-be investors that expected returns are also following, leading to a decreased desire for capital investment."

It sounds like you understand that, but are arguing that the causality is reversed, which seems a very legitimate critique.

Comments for this post are closed

Apologies, the rephrase is not indicative of the authors' claims.

Comments for this post are closed

Comments for this post are closed

In a fiat currency regime, inflation rates and interest rates are a decision by the central bank. The Western Central Banks have, for whatever reason, decided to drive inflation down to almost zero and so interest rates have also fallen. It has nothing to do with demographic shifts or slower growth, other than the fact that an aging population might want low inflation and so vote for politicians who promise to implement it.

If you don't believe that Central Banks can generation any desired inflation rate, please lobby your congressman to get the Fed to buy up and retire the entire national debt of the US, and additionally fully fund everyone's retirement account with $1bn by printing money.

Great points. Obviously, the world central banks have expressed a nearly hysterical squeamishness about inflation during the last 20 years.

That said, there can also be a lot of savings, due to sovereign wealth funds, demographics, compelled saving such as public pension plans, and even insurance regulations.

The real question is, why does anybody believe interest rates will rise in the next 10 to 20 years? And so why does the Fed talk about normalizing rates?

Comments for this post are closed

Real versus nominal interest rates. Big difference. Central banks cannot control real rates, only nominal ones.

As an empirical matter, QE raised real interest rates. That one blows people's minds. It's almost like a bolder Fed inspired more confidence among the populace or something.

Can you expand on this? For instance, here is the FRED chart for 10YR TIPS. But even this is biased by the fact it is based on market forces which include inflation expectations.

The previous link requires chart transformation. This should be a clean chart of the TIPS.

Comments for this post are closed

I'll take a crack at it.

TIPS yields peaked in October 2008 and began falling fast.

The Fed initiated QE1 in December 2008 in this environment. Rates continued to fall. The Fed doubled QE1 in March of 2009, at the trough of the financial crisis. TIPS yields bottomed out around 1.28% in March of 2009.

Between March of 2009 and March of 2010, when QE1 ended, yields fluctuated, but overall they rose to 1.68% by March of 2010.

With the withdrawal of QE, yields plunged, bottoming out around 0.50% in October of 2010, which coincided with the start of QE2.

Yields jumped to 1.34% in March of 2011, before starting to slide again (QE2 was weak tea). Still, yields were at 0.80% in June 2011 when QE2 ended.

At which point, yields tanked. going negative by August 2011.

At this point, the Fed shifted gears, implementing Operation Twist in an effort to flatten what was a steep yield curve by swapping short-term bonds for longer term ones. I think it achieved it's objective, but this was not QE.

Yields continued to sink throughout 2012, reaching -0.80% in September of 2008, when the Fed announced QE3.

Markets were skeptical at first, so the Fed stepped on the accelerator in December of 2012 (WE REALLY MEAN IT), and by June of 2013, TIPS yields had popped back up into positive territory around 0.50%.

Then the taper tantrum. Could the Fed delicately move away from QE without yields tanking? Well, yields have bounced around since then, flirting with zero in 2015.

And then look at recent activity. In December, the Fed unquestionably tightened- the opposite of QE. TIPS yields have fallen about 0.25% since.

Comments for this post are closed

Thank you. How do you think your interpretation relates to Nathan's statement that "Central banks cannot control real rates, only nominal ones"?

Comments for this post are closed

Tips are inflation adjusted bonds, where the principal is adjusted by inflation.

The yield for tips bonds depend on the inlation rate and the demand (or market price) for bonds. Did the yield change as a result of inflation expectations (investors buying due to expected inflation when the fed prints money) or due to actual inflation?

Post 2008 there was a wide feeling that stimulus and fed action would cause inflation, no doubt prompting high demand for tips bonds. And when the expected inflation didn't happen the demand dropped off.

This confirms my understanding that the fed mostly maintains an illusion of omnicience by inciting behavior in the market as opposed to actual buying and selling large enough numbers to move the market. And that their primary policy objective is to maintain that illusion. They lost it in 2008 and had to buy trillions of dollars worth of assets and print even more trillions to keep the banking system working.

Comments for this post are closed

Well, 'control' is of course way too strong, but 'influence' is not.

The Fed's influence on real rates is indirect and complicated.

Comments for this post are closed

Reading this paper in detail, it is very much my old argument, that the Fed is the second driver on the hook and ladder truck. See chart B2

Comments for this post are closed

Comments for this post are closed

Nathan W and Brian - "real" interest rates are comprised of two elements, a factor for return on your money and a factor to account for unexpected inflation. If inflation is expected to be low and remain low then real interest rates will be low. So when CB lower inflationary expectations, which they have been doing by stamping out every hint of potential inflation, then they lower the risk factor for unexpected inflation as well. So real interest rates will fall along with nominal ones.

Comments for this post are closed

The real rate is defined as the nominal rate minus inflation (econ101, we all know). I wouldn't be surprised to find that the observed real rate according to the definition could fluctuate in response to monetary policy in the short term, which should be very useful in fighting a recession. But I would be surprised if even a perfectly coordinated move by every central bank on the planet could affect the real rate in the long run. Perhaps this is a dogmatic view, but I'm inclined to believe it. Narrowing in on short-term empirical evidence is just clouding the matter, according to this view.

It seems you are quite more the authority on this question than me though :)

Nathan - it is fairly easy to calculate what the real rate you have received, as you say it is the nominal rate minus actual inflation. But nominal interest rates are not set by historical inflation, they are set by expectations on future inflation. So if you as a lender observe that the central bank does not have good control of inflation, like for instance in the 1970's, you have to include a factor for further inflation growth, beyond what you already see today. If inflation in fact falls (as it did), lenders will then receive a windfall reflected in high calculated real rates. Now however everyone sees that inflation is well controlled and actually perhaps going to fall further. The recent rate rise by the Fed is further proof of this, no inflation but the Fed raised rates. So nominal interest rates reflect that expectation. It may even be that this "unexpected" inflation factor is now negative. So people are betting that we will more likely have deflation rather than inflation and lowering their expectations accordingly.

So it is the case that the CB can affect both nominal and real interest rates. High inflation coupled with concerns over further high inflation will give you a high nominal rates, and the real rate of return will appear high when calculated using today's inflation rate. Low inflation and deflationary expectations give you low nominal rates and the real rate will appear low when calculated using today's inflation rates.

Comments for this post are closed

"nominal interest rates are not set by historical inflation, they are set by expectations on future inflation"

It seems obvious now that you say it. Hadn't crossed my mind. I hadn't understood how the real rate could fluctuate in the short term. I'm still doubtful that this could affect real rates in the long term.

Comments for this post are closed

Nathan - here is a chart of real interest rates in Zimbabwe as they rocketed to hyperinflation. At the end the real interest rate was 600%. Still don't believe inflation rates have an impact on real interest rates?

Comments for this post are closed

I don't often see an instance of a contrary example as evidence that a general theory may generally hold. But that is certainly strong contrary evidence.

The size of the informal dollarized and randized economy might mean that the official data could be misleading. For example, I remember getting ten times the Zimbabwe dollars on the black market as I would have gotten at the bank exchange rate - even banks had a special exchange booth for the information transactions - wealthier tourists who were paying for everything on a credit card were certainly getting fleeced at the time (2001, inflation was very high, but not yet hyperinflation). Presumably most loans were in dollars and rand, as the value of the Zim dollar was volatile and essentially unpredictable (generally falling fast). Zimbabwe dollars were basically used to pay government salaries, which I believe was basically the means by which cash was extended into the economy. It was utterly impossible to exchange any remaining Zim dollars outside of the country, but there were numerous informal currency traders at the border, as is normal at many borders in the world, giving people a last chance to exchange their soon-to-be worthless paper before leaving the country. Zimbabwe might be a bit of a special case for these and other reasons.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Central bank manipulation definitely caused these abnormally low interest rates.

Such interventions distort the entire structure of asset prices and penalize/tax savers. Low nominal interest rates, even at low rates of inflation, can mean negative real rates. When interest rates are held too low, asset bubbles develop and investment funds are misdirected to unproductive uses; painful economic corrections always follow.

Monetizing government debt and controlling interest rates are long failed experiments in market socialism. Of course, the core purpose of central banks is to buy government debt so that government can continue fiscal profligacy.

Most economists still cling to Keynesian aggregate demand management as the miracle cure for a stagnant economy... believing central banks bureaucrats can actually stimulate real economic growth and lower unemployment by inflating the monetary base and prodding consumption. Thus, we hear that letting inflation exceed 2% could have social benefits that far outweigh the real costs, that consumption needs to be pumped up, that saving is harmful because it cuts aggregate demand, and that artificially low interest rates stimulate investment. Following Keynesian doctrine leads to economic decline... as should be obvious in observing the current U.S. and world economy.

Klausner - Can you identify these asset bubbles that are developing for me so I can become as rich as you by shorting them when they pop?

But given you believe that bubbles are caused by too low interest rates, you must also believe the corollary, that economies can be depressed by too high interest rates. I mean this statement is just the mirror image of the first surely. In which case we are in agreement that the Fed has a role in managing the economy, since they set interest rates. So given we both acknowledge the role of the Fed now we need to decide on a target for the Fed. You suggest that inflation above 2% is not desirable, I agree with that, so I guess we can set the Fed the target of 2%. Or do I misunderstand you?

It is a fluid model, and the new money enters the system and is distributed unevenly in millions of complex transactions so the effects quickly become untraceable. We only know where the unsustainable activity was after its liquidation.

Having said that, I was skeptical of positive rates of return on bundled mortgages including $500K fixer-uppers in East LA. And some fund managers made a killing betting against their own clients' money in MBS's.

Some others to watch: derivatives on bundled auto/pawn-title loans (yes, they exist); student loans to students who have no business going to college; Canadian real estate. Of course, if your country's central bank is going to be the ultimate buyer of last resort, then you might want to dive in.

This is the problem with central banks. They are distorting market signals so nobody knows what assets are worth.

AG - its a fluid model sure - of course the world is complicated - but what do you want the Fed to do? Keep interest rates the same as now for ever? Lower them? Raise them? What indicators will they follow to decide if they are succeeding or failing and need to adjust rates? Keep in mind that we have agreed they can have a significant effect on the real economy.

Comments for this post are closed

"– but what do you want the Fed to do? "

----go away, vanish, leave us alone

Fed bureaucrats have no special abilities
they don't know what interest rates 'should' be---- any more than they know what the prices of door knobs, carrots, or cruise ships should be

Comments for this post are closed

@dwc - I am not exactly sure what you mean when you say you want the Fed to vanish. Do you mean to keep the current dollar based economic system, or do you mean a private money system with state currencies replaced? If you mean the first, we still have to decide how money supply will be regulated - like for instance keeping the stock of currency fixed at the time of Fed abolishment. Or perhaps linking the currency to gold. Or perhaps adjusting the money supply to keep inflation constant. Please tell me you thoughts on this approach and the pros and cons of these options. Clearly we would want to have this well thought through since the entire economy is at risk if we get it wrong. We saw a gold based currency in the 1930's cause lots of economic harm and lead to the election of a left wing president that ended up with some very bad results for us libertarians and we don't want that again.

On private currencies, they will also face this same problem of deciding how to regulate their supply of money to their users. As an employee I would like my salary in real terms to be consistent with the cost of living, I would not want one week to be able to afford steak, and the next week be living on beans because my salary in nominal terms remained the same, but in real terms depreciated. I would also see that it would be good if my salary didn't appreciate rapidly in real terms, then my employer would probably not be able to afford me. So I would suggest that any private money supplier would end up in the same place as the Fed right now, i.e. regulate the money supply to ensure that in real terms it remains broadly constant. There is also an additional factor that I, as a private money supplier would have to consider, which is that if I set the value of the currency to remain at exactly zero, any small deflation factor pushes rates negative, which is not a good thing for any banking system, why save in bank if they are going to charge you?. So I would add a small inflation factor to keep things away from this zero lower bound. Hence I, as a private money supplier, would be regulating my money supply to achieve an inflation target of +/- 2% give or take or basically where the Fed are now. What would you do if you were the private bank?

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

At first sight it seems promising. At second sight they do not define ´investments´. Implicitly, looking at their graphs, their investment defintion clearly includes houses. They do however not specify their source for the price of investments other than (IMF 2014) and IMF 2014 contains a vague reference to Gordon 1990. Sigh.

In all probability their price series for investment goods does, in bad neoclassical tradition, not include houses. The source they use is Eichengreen (2015) but Eichengreen does not mention which NIPA series on prices he uses. In all probability the series is restricted to market organisations and excludes households and the government which means that it excludes roads, government buildings, many harbours, bridges and houses.

Quantitatively, houses (including land underlying houses) are the most important kind of capital, followed by (buildings plus roads). I do admit that prices of those declined in Japan. But did they decline that much in ´The West´? Looking ahead: in the next sixty years 5 billion additional people will need new houses, and another few billion will need better houses. I´m not entirely convinced about these falling prices of investment goods.

By the way. A better long term USA investment series than the one used by Eichengreen (as well as series for a large number of other countries) is available here: http://www.paecon.net/PAEReview/issue69/Knibbe69.pdf

While houses are treated as an investment by many owners, economically speaking they are consumption. Housing can be reasonably be regarded as investment when referring to improvements from a tent, for example, because human capital is protected when people are protected from infestations of disease-carrying insects and rodents, protected from illness due to extreme heat, cold, rain, etc. But at our standard of living, housing is basically consumption, not investment, for economic purposes. It may serve as a store of savings, but only by virtue of the fact that someone else might pay lots of money for it, because they wish to consume accommodation.

In short, housing cannot be used to produce other goods (yes, you can partition off some part and use it for work, but then this share of the building is not considered as "housing"), and therefore it is not really "investment".

However, perhaps there are good reasons to nevertheless consider it as investment for various forms of economic analysis, especially relating to financial market.

Comments for this post are closed

Comments for this post are closed

It seems to me that, eventually, the real interest rate cannot be higher than some addition of the rate of return on natural capital plus productivity growth. In an era where major gains could be made by exploiting non-renewables this did not apply. Not sure how to defend this opinion in a formal theoretical structure though. In an economy which is 100% based on exploitation of nature (hunting, fishing, savenging), the rate of return on capital is clearly the maximum bound for the long-term discount rate. In the modern economy there are other forms of returns, but it is not clear to me whether this implies higher or lower real rates in the long term than the rate of return on natural capital (regeneration rate).

The factors mentioned in the cited paper seem relevant perhaps to understanding fluctuations around such an underlying fundamental discount rate (conceptual), but I don't agree that they are themselves the driving factors in overall change. Rather, the eventual end of exploitation of non-renewables and technological barriers which slow productivity growth (unproven) should be the factors driving the actual fundamentals of the real rate.

Weitzman's survey of economists' opinions on the correct long-term discount rate (http://scholar.harvard.edu/files/weitzman/files/gamma_discounting.pdf) suggest that a lot of economists think this might be in the range of 1-5%, and stunningly from his survey, there are people employed in the economics profession who peg the real discount rate as greater than 20% and others pin it as negative! (a very small minority, however). I would be curious to see a repeat on such a survey in light of recent developments in the global economy.

Comments for this post are closed

And therefore all pension funds are bust.

Well, it makes the process of financing retirement more difficult and expensive, including 401(k)s.

Private pensions in the US (close to $3 trillion) are basically solvent. Under US and IAs rules, pensions have been measured based on market interest rates for decades. Employers got the memo early, and the majority of private sector pension plans have been in the process of winding down for years now, but it's a long tail and, in aggregate, companies are still a couple hundred billion in checks away from wrapping this up.

A minority of employees continue to maintain pension plans despite the stringent requirements. These plans are fine.

In DC plans, employees are getting the meme and responding in obvious ways (save more, work longer.)

The unsolved problem in the US are the ongoing public sector pensions, which continue to use fanciful 8% numbers in their calculations and often can't be shutdown. A recent law in Illinois aimed at reining in future pension accruals was struck down last year by the Illinois Supreme Court.

I suspect in Europe, where state pensions play a larger role, the situation is more dire. The memo has not gotten through.

Regarding Illinois, the Democratic Party has controlled the government for a decade and is in thrall to public sector unions.

A state rep that broke with the party on pensions is now being destroyed with old tapes, none of which were apparently an issue during the 12 years when he toed the line.


Anyway, this guy may have 100,000 voters in his district, and I think 15,000 families in the district have government jobs, so he's gotta be dead in the water, right?

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Not convinced.

The authors list some factors that should increase the savings rate if the return on invested capital stays constant. Then they observe no change in the savings rate. Instead of scrapping their paper and going back to the drawing board, they just assume that the investment opportunities have also worsened. That's my reading of the paper anyway.

How about my explanation: the very long term government bonds were very underpriced in 1980 because people didn't have much trust in governments. Especially in terms of inflation, but also the governments willingness to pay back the borrowed nominal monies in full Instead of taxing them away.

As a minor point, how about looking at the real after tax rates, given that nominal interest payments were taxed at rates that were over 50% in the US and over 90% in UK in the 1970's and inflation was high. Do the real rates still look like they are trending down? LOL as my daughters would write...

Comments for this post are closed

Broadly speaking, that's what I said.

Comments for this post are closed

Remember, my favorite chart covers 160 Years Of Interest Rate Fluctuations For 9 Major Countries

I didn't think you could get 160 years of what I called "flocking" because generations of central bankers were dedicated to that task. They had other, diverse, priorities. You get it instead if it comes out of the global system itself, or as these authors say:

"The co-movement in rates across both advanced and emerging economies suggests a common driver"

That co-movement goes back a long time.

A counter argument to myself at BI


If one believes the world moved to synchronized monetary policy in the early 80s then one might believe that is sufficient driver for the synchronized fall to negative rates.

Still can't quite believe that these central bankers at once failed to meet their own goals, ended up someplace undesirable, all together, because they were in control of their destiny.

Comments for this post are closed

I see that my favorite chart is their Chart A1.A

Comments for this post are closed

Comments for this post are closed

Yeah, demographics.

Overlay that with the financial scars suffered this century (investors have been mauled, because the Fed is just out for rich people or something), producing a mindset that makes GUARANTEES OF ANY KIND VERY EXPENSIVE right now.

I think it's actually a great time to invest in risky assets. Because so many people remain fearful. Nobody was afraid in 1999.

If your last sentence is correct, then people deep pockets who can withstand high volatility in the short and medium run are poised to make an awful lot of money, just like they usually do in such situations. Those who plan to retire on a modest income in 10 years are probably the biggest losers of such a situation.

Yeah, I think that's right.

To me, the most important asset an investor possesses is time horizon. When will I need the money?

If the answer is "in less than five years", you're not really an investor. In the current environment, it'll be hard to do better than preserving purchasing power in this case, without running the risk of a significant loss.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Somewhat OT:

This is from the FRED TOS. Are there any lawyers that review the terms and indicate if we should be aware of any other important changes?

You will not use (or permit the use of) the graphs or charts in any manner which is harmful, threatening, unlawful, defamatory, infringing, abusive, inflammatory, harassing, vulgar, obscene, fraudulent, invasive of privacy or publicity rights, hateful, or racially, ethnically or otherwise objectionable;

The SJWs are stopping you from putting a Confederate flag on a Phillips curve.

LOL. So long as that remains the definition. I missed bolding the word "harmful" in that.

Would posting a chart breaking down unemployment by race be considered "racist" while posting a chart breaking down home ownership by race isn't?

Did you know there is a push (maybe enacted legislation, unsure as I do not follow it) for considering sex between two intoxicated adults as rape? Maybe they blood test them after and assign sentencing based on BAC, like person A is .05 and person B is .10; so A get .33 of a normal sentence and B gets .67 of a normal sentence. Hooray for America.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

Recent real rates are actually quite normal. Of course you need to use private credit rates, not sovereign rates.

Doing a simple comparison on Fred of Moody's AAA vs annual CPI, what I see are three well-known effects:
1) real rates rise as inflation rises, but with a small delay
2) real rates fall as fall inflation falls, but with a big delay
3) real rates are cyclical, peaking during recessions and troughing in late expansion.

Of course it's possible to force other phony conclusions out of the data, eg by comparing recent real rates to the early 1980s, when recession and a recent inflation spike led to a historic peak of real rates.

Not those rules all apply to measured, backward-looking real rates. Forward-looking real rates are another story.

Ooops, I meant to write "Note those rules all apply ...".

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

"...not least in how to manage the business cycle if monetary policy is frequently constrained by the zero lower bound."

This is why we should consider a fiscal counterpart to interest rate management, such as a VAT or national sales tax (especially if it's combined with sensible offsets like lowering corporate tax rates in the US). If allowed to flex---as interest rates are---in the hands of skilled appointees, it could offer valuable steering potential. Raised in times of an overheated economy or lowered in a stall, we might gain the equivalent of a fall-back economic seat belt...one that remains functional near the zero bound of interest rates.

Comments for this post are closed

That "zero lower bound" is being thrown on the ash heap of economic history by the Japanese and Swedes, et al.

Comments for this post are closed

Comments for this post are closed