Two big mysteries

Nick Rowe has a post discussing 30 year bond yields in Canada.  He begins by quoting from an official at the Bank of Canada:

All told, we think that the neutral rate of interest is lower than it was in the years leading up to the crisis because of these structural developments. We estimate that the real neutral policy rate is currently in the range of 1 to 2 per cent. This translates into a nominal neutral policy rate of 3 to 4 per cent, down from a range of 4 1/2 to 5 1/2 per cent in the period prior to the crisis.

Then Nick comments:

But what puzzles me is this: the 30-year bond is currently yielding 2.74%. That’s below Carolyn’s 3% to 4%. And Carolyn is talking about the Bank of Canada’s target for the overnight rate (the “policy rate”), and that’s normally below the 30-year bond yield.

I checked the 30-year rate in Germany and the US:

US 30-year bond yield = 3.21%

German 30-year bond yield = 1.91%

So it’s a sort of global mystery.  Why are long-term interest rates so low?

But it’s not the only big global mystery.  There’s also the absurdly slow “recovery” from the 2008-09 recession.  This is the only recovery I ever recall seeing where the growth rate of real GDP in the US was below the trend growth rate.  And nominal GDP growth has also been very low. Indeed both growth rates might well be the lowest in American history for a recovery period.

And it’s not just the US.  Some countries like Canada may have done a bit better, but lots of countries in Europe are seeing even slower recoveries in both NGDP and RGDP.

I am always suspicious of coincidences.  I suppose you could dream up an explanation for the low bond yields.  Maybe all that QE has depressed long term bond yields.  Maybe I am wrong that money has been really tight; maybe it’s really easy, and that explains the low bond yields.

But there’s one problem with that explanation–it makes the other mystery even more mysterious. How likely is it that money has been so easy that it has produced almost unbelievably low bond yields, and yet NGDP is growing at the slowest rates ever seen in a recovery?  Easy money is supposed to lead to fast NGDP growth.  I don’t like “answers” that solve one mystery at the expense of making another mystery even more mysterious.  [There’s no law of the conservation of net mysteriousness.]  I like “answers” that solve both mysteries at the same time, using standard off-the-shelf economic theory.

The Great Stagnation!

It explains the low bond yields, and it explains the slow economic growth despite rapidly falling unemployment rates.  But what explains the differences in bond yields between countries?  Here I’d like some help, but I’ll throw out a couple tentative hypotheses.  Although both the US and Canada have 2% inflation targets, I recall reading that the US is more aggressive in the “hedonics” of price adjustments.  Indeed one Fed official said their 2% PCE target is actually more like a 2.4% CPI inflation target.  Is that right?  If so, might that explain why Canada has lower bond yields?  If both countries have 2% inflation targets, but Canada measures inflation more conservatively, then Canada would end up with lower NGDP growth, other things equal.

In the case of Germany, it might be related to the fact that the ECB is targeting inflation at slightly below 2%, an asymmetric target.  In addition, inflation has recently fallen to only 0.3%, and the ECB’s response — how can I put this politely — hasn’t exactly inspired confidence.

So perhaps the entire developed world is entering the sort of growth stagnation that hit Japan 20 years ago, producing low real interest rates everywhere.  Then the differences in bond yields reflect differences in the techniques used to measure inflation, differences in inflation targets (especially the ECB), and differences in how confident investors are that the central bank will actually hit its target (the ECB and Japan.)

I’m glad to see the BOC re-evaluate their interest rate assumptions, but the markets suggest they are still behind the curve.  Of course the Fed has consistently overestimated the Wicksellian equilibrium interest rate, and hence over-estimated RGDP growth, ever since 2007.  And the ECB? There aren’t any words to express my contempt for that organization.

We’re 15 years into the 21st century.  How much longer will it take the world’s central bankers to realize that we aren’t in Kansas anymore?  None of the 20th century rules of thumb have any value in today’s world.



45 Responses to “Two big mysteries”

  1. Gravatar of Jason Smith Jason Smith
    25. September 2014 at 18:59

    I think it’s all about a future of chronically under-shooting not only inflation targets, but even inflation expectations … This same picture applies to the Fed, ECB and BoJ

    In the model, as economies grow large, the grow “cold” more likely filling all the low growth markets (by random chance).

  2. Gravatar of Jason Smith Jason Smith
    25. September 2014 at 19:01

    (This is essentially a model of the great stagnation, except you can escape it by resetting what money is/means)

  3. Gravatar of Kevin Erdmann Kevin Erdmann
    25. September 2014 at 19:07

    There is an ever-changing maturity premium that we tend to think of as positive. But I wonder if global demographics are stark enough to actually create a negative maturity premium.
    There are a heck of a lot of people around the world who will be unproductive after about 2030 but want secure income in 2050. This premium clearly fluctuates, so there isn’t a fixed premium that arbitrage pushes to. Why couldn’t that premium go negative under extreme conditions, like when a world wide war creates a demographic hump?

  4. Gravatar of Saturos Saturos
    25. September 2014 at 21:34

    Scott, have you seen:
    1. New Michael Woodford interview:
    2. New Narayana Kocherlakota speech:

  5. Gravatar of Lorenzo from Oz Lorenzo from Oz
    25. September 2014 at 22:11

    Not the entire developed world. Australia’s NGDP growth rate has recently returned down to trend. Just saying.

    We apparently do not do 30 year bonds (at least, the RBA website does not report them). But our 10 year bond yields are declining.

    Although the current 10 year bond yield is 3.495.

  6. Gravatar of Dave Dave
    25. September 2014 at 22:38

    Thinking out loud, Australia may be the exception because it’s one of the few developed countries with an inflation band as opposed to a target, which permits more flexible policy and kind of approximates NGDPLT. So targeting regimes have led to more pessimistic long term growth expectations.

    If you’re concerned about hedonics, just look at current price GDP in industries with less reliance on things like semiconductors/computers.

  7. Gravatar of Roger Sparks Roger Sparks
    26. September 2014 at 05:01

    Why don’t we simply admit that the Central Banks DO control interest rates, including long term interest rates? The so called “neutral rate of interest” may (simply) not exist.

    My understanding of the loop-of-government-borrowing, central banks, and private holders is that governments borrow from, but not directly, from central banks. Instead of direct borrowing, governments must first borrow from private sources. Then later, the private sources sell government bonds to the central banks. The private intermediate holders always participate at a profit which results in bonds selling at higher price than purchase price. Higher price in bonds results in lower yield (so the average yield is reduced in this process).

    It seems to me that if we want to investigate a “neutral rate of interest”, we must take central banks completely out of the ownership loop.

  8. Gravatar of Matt McOsker Matt McOsker
    26. September 2014 at 05:27

    Long term bond yields are low because expectations for future growth and inflation are low. We are in a secular, global deflationary cycle. Maybe deflation is too harsh, so let’s call it low inflation.

    Global monetary policy for short rates is set on low, AND fiscal policy is weak. Short rates are low because central banks have set low rates to try to spur growth (not saying this has not worked, but may have limits). The growth remains weak. so rates stay low, and Central banks are using different methods to try and get them lower (e.g. negative IOER).

    Long bond yields are priced based on short rates, current inflation, and future inflation expectations(maybe next 2-3 years).

    There is a lot of idle labor, and there is a global surplus of productive capacity. Is there any doubt that if demand for cars or other goods increased, that demand could not be me? Add it all up, and it is no surprise long term yields are low, and growth has been poor.

  9. Gravatar of Nick Nick
    26. September 2014 at 05:44

    What does ‘the great stagnation’ mean to you? As cool as japan is, it’s hard for me to imagine that it reached the frontier of technology ten or twenty years before the US. If population growth and monetary policy are still factors #1 and #2 in explaining low long term yields what does ‘stagnation’ add to the picture?

  10. Gravatar of Steven Kopits Steven Kopits
    26. September 2014 at 05:51

    If you check, you’ll find that pretty much every country has underperformed the IMF’s 2010 GDP growth expectations.

    The exceptions:

    – Canada (about par)
    – Israel (a bit above par)
    – Saudi Arabia (well ahead of expectations) I haven’t checked the other, stable Gulf OPEC countries, say, Qatar, UAE or Kuwait. Want to guess how they’ll rank compared to 2010 expectations?

    So no one wants to invest? Why doesn’t this apply to the oil business? We’re seeing record capex again this year, against some incredibly bad free cash flow numbers from the majors (see my post: Aside from shale oil, this is an industry that’s not delivering and yet capex continues to pour in. Why is that? Are wages in the oil business low? Where is the secular stagnation in oil? It’s one continuous boom. What does not tell us about the global need for oil?

    So why not preface your comments with, “Aside from the oil business, where upstream capex is rapidly approaching $1 trillion per year, investment in other sectors, particularly in cities producing oil consuming goods–like Detroit–is depressed.” This is not a depressed global economy. It’s a depressed global economy in pretty much every other sector except oil. If secular stagnation were the true cause, oil prices would be weak. And they are not. Even with 2 mbpd / year of production increase from North America (astounding!), a feeble Europe and an uncertain China, oil prices are still close to $100. That’s just incredible, and completely argues against any notion of some depressed global economy. Instead, it argues that we have a constrained global economy, that we are short on a critical resource.

    Let me add an interpretation related to John Fernald’s recent paper for good measure. He refers to the pre-Recession “bubble-economy sectors of natural resources, construction and real estate, and finance.”

    Let’s see what blew up of those. Construction did. Real estate did. Finance did. Oil production did not. It was the scarce resource and remains the scarce resource. It’s the binding constraint, and it has to be. If you understand the implications of the FCF chart I link above, you’ll appreciate that oil has to be the binding constraint. My prediction from three years ago would not have worked without it.

  11. Gravatar of Student Student
    26. September 2014 at 06:23

    The great stagnation hypothesis may explain the simultaneity of low bond yields and NGDP growth but it presents a huge problem in my mind. That being it contradicts growth theory and so is a mystery in its own right. Investments in human capital/knowledge have increased, barriers to spillover overs have declined dramatically, and yet we have slowing innovation? That doesn’t make any sense.

    Further, I dont agree with all of the pessimists out there arguing innovation has even slowed down. It took thousands of years to get to flight, 50 years to get from flight to space, 10 years to get from space to the moon, and 40 years from the moon to Mars. That’s one little example but it doesnt seem to jive with a slow down in innovation as i see it. Further, robotics are exploding before our very eyes, as is artificial intelligence, access to information (information that took weeks to collect and much expenditure can now be obtained in minutes at close zero cost… spillovers again), 3d printing, materials science, on and on. Innovation has not slowed down, far from it.

    Basically, TC’s low hanging fruit solution is quite problematic IMO. There is plenty of unused land here on earth and the moon, mars, and near earth asteroids are pretty thinly populated (although that is far from free at this point). There are plenty of returns to education remaining (what would happen if public education went through a masters degree or so?). Solar costs are plummeting (a pretty available source of energy), and the american constitution still exists. As well, putting rapid invention on the right hand side of the equation (a source of low hanging fruit) is a serious endogeneity problem in the first place IMO.

    Solutions that are themselves mysterious, dont really seem to solve anything.

  12. Gravatar of Brian Donohue Brian Donohue
    26. September 2014 at 06:40

    Demographics? It’s The Great Age-nation.

  13. Gravatar of Effem Effem
    26. September 2014 at 06:57

    Loose monetary policy lacking a transmission mechanism out of financial assets and into the physical economy would also answer the paradox.

  14. Gravatar of Doug M Doug M
    26. September 2014 at 08:41

    The difference between NGDP and RGDP is the GDP deflator.
    The deflator is not a hedonic index.
    So, while CPI may be understated because of aggressive hedonics, this does not mean that Real GDP will be overstated.

  15. Gravatar of ssumner ssumner
    26. September 2014 at 08:51

    Jason, The data does not support that. Nick has other posts showing that Canada is hitting its inflation targets. And it doesn’t explain how the unemployment rate could be plunging rapidly while the economy is growing at 2%.

    Kevin, Possible, but I doubt that can explain it. And of course it doesn’t explain the second mystery.

    Saturos, thanks for the links.

    Nick, In the UK RGDP is no higher than 6 1/2 years ago, even as employment is hitting records each moth. Neither money nor population can explain that. The US case isn’t quite as stark, but the same sort of process is playing out here.

    Having said that, money and population do play a role.

    Matt, But demand-side factors cannot explain the rapidly falling unemployment rate, combined with 2% RGDP growth.

    Lorenzo and David, Australia’s always had a higher trend NGDP growth rate, and higher nominal interest rates than other developed countries. But haven’t both slowed somewhat in recent years?

    Roger, You said;

    “Why don’t we simply admit that the Central Banks DO control interest rates, including long term interest rates? The so called “neutral rate of interest” may (simply) not exist.”

    We don’t admit it because it’s not true. A shift toward easier money often makes long term yields rise.

    Steven, Didn’t you tell me 6 months ago that capex in oil was falling? In any case, in recent decades oil is no different from health care–lots of growth in gross output, horrible productivity numbers (except for shale, obviously.) Productivity is stagnating in oil as in the overall economy. If you want an exception it’s smart phones.

    Student, You said:

    “That being it contradicts growth theory and so is a mystery in its own right. Investments in human capital/knowledge have increased, barriers to spillover overs have declined dramatically, and yet we have slowing innovation? That doesn’t make any sense.”

    That’s not “growth theory,” those are your assumptions, which I don’t accept. Investments in education might have reached the point of diminishing returns, even near zero returns at the margin.

    You mention aviation. Here are state of the art airplanes:

    1902- Wright brothers.

    1927 – Lindbergh crosses the Atlantic

    1969 – Boeing 747

    2014 – Boeing 787 (747 is still Boeing’s biggest)

    You don’t spot a downshift in technological progress? I do.

    You said:

    “There are plenty of returns to education remaining”

    I doubt it. The problem with education is that students don’t study (except Asian-American students, of course.)

    Plenty of land? NIMBY rules prevent its use.

    Regarding all your comments on AI etc, that just shows that the Great Stagnation will end at some point in the future, and robots will take off. Even Tyler agrees with that. But last time I looked people were still driving their own cars and cleaning their houses with mops.

    Dude, Where’s my Flying Car?

    Effem, You can’t explain long term stagnation with demand-side arguments. And you can’t explain why unemployment is falling rapidly as RGDP growth is 2% with demand-side arguments.

  16. Gravatar of benjamin cole benjamin cole
    26. September 2014 at 09:08

    Excellent blogging. Print more money, cut FICA taxes.

  17. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    26. September 2014 at 09:44

    It is fruitless to ponder WHY interest rates are what they are. 33 years ago it was abnormally high rates that even Milton Friedman couldn’t explain;

    Current high interest rates offer a puzzle for which, so far as I can see, we have no satisfactory explanation. While many
    explanations are glibly offered, all of them are contradicted by past experience.

    Since the mid-1960s short-term interest rates have tended to move up and down both with current inflation and, after a lag, with ups and downs in monetary growth. Inflation has been declining for some time now, despite the isolated sharp rise in the consumer price index for the month of July. Similarly, monetary growth declined moderately from April to May and has been relatively stable since. Yet interest rates have stubbornly stayed at record levels.

    Clearly, this needs explanation.

    And there has been no shortage of proposed explanations. The trouble is, all of them are wrong.

    He then goes on to list three such explanations and why they are wrong, or contradictory. But he admits that he doesn’t know what the correct explanation is (was).

    ‘The failure to find a reasonable explanation does not mean that there is none.’

  18. Gravatar of Matt McOsker Matt McOsker
    26. September 2014 at 10:42

    Scott writes:

    “Matt, But demand-side factors cannot explain the rapidly falling unemployment rate, combined with 2% RGDP growth.”

    The UR has come down, but partly due to falling participation rate. Do you feel the 2% RGDP is low compared to a rapidly falling UR? I would agree that it is, but seems to be several explanations:

    – UR is coming down largely due to part rate
    – PCE component of GDP has been steadily climbing so there is some positive growth there
    – Trade deficit has not been as bad as it was prior to the recession, but recently climbing and acting as a slight drag
    – Domestic investment has been steadily climbing a positive
    – Government consumption expenditures has been flat in nominal terms since Q3 2010, and has dropped slightly in real terms – which is probably the main drag on GDP. Even government employment is down by something like 500,000 since the recession.

    So it seems there are a few factors shaving growth off of GDP, and some explaining the falling UR. Then add to to that credit factors. There is enough juice to keep growing, but at what many feel is too slow a rate. Most of the above is specific to the U.S. But, IMO we have global monetary and fiscal connections as never before.

  19. Gravatar of Todd Kreider Todd Kreider
    26. September 2014 at 10:44

    There are serious problems in arguing a major slowdown in aviation innovation. Tyler Cowen uses the same example by showing a picture of a plane circa 1970 and one from 2010. They look similar, but significant changes have dramatically lowered prices so that not only rich economics professors can fly. The accident rate has also dropped dramatically since 1970.

    I forget the exact number but if one compares the death rate of flying per mile in the U.S., around 1940, it would be as if a major airliner crashed and killed 300 aboard… every day.

  20. Gravatar of Brian Donohue Brian Donohue
    26. September 2014 at 11:11

    @Patrick, I threw out demographics as a common thread to the mysteries Scott identified. I’m sure this is part of the explanation- it nicely identifies Japan as the canary in the coal mine.

    Doesn’t this also help explain the high rates of the 1970s? Lots of young Boomers plus Friedman’s own permanent income hypothesis.

  21. Gravatar of Student Student
    26. September 2014 at 11:46

    Scott, you said “That’s not “growth theory,” those are your assumptions, which I don’t accept.”

    Economic growth being driven by investment in human capital and the existence of positive spillover effects that reduce diminishing returns to capital accumulation are not my assumptions, rather the basic tenements of economic growth theory per Romer, Lucas, Aghion and Howitt, and Grossman and Helpman.

    Also, the time from 1902 (first flight) to 1969 (modern commercial aviation) was 67 years. The time from sputnik (1957) to the first landing on Mars (2012) was 57 years. I dont see the technological downshift there. To me, its a much bigger leap to go from the first satellite in space to the first landing on another planet than it is to go from the first flight to a 747… and it took less time.

    And for the record, a 787 is not a good representation of the state of the art in aviation in 2014. I would say a venture to guess the state of the art is some sort of unmanned aerial vehicle that is undetectable by radar or heat sensors, produces no sound, and is capable of interacting with other unmanned aerial vehicles independent of human intervention… but only the CIA would probably know that for sure.

    I completely agree that the problem in education is that many students dont study (btw, i’ve seen pleny of asians that dont and plenty of non-asians that do). However, imagine what would happen if we could motivate them to? Imagine if we paid thousands of kids to study online rather than paying all that money for brick and mortar buildings. Diminishing returns… I think not.

    Flying cars… you’re so 1950’s jetsons

  22. Gravatar of Steven Kopits Steven Kopits
    26. September 2014 at 13:01

    Scott –

    Capex is falling at the oil majors. It appears to be rising for the shale players. I have to get hold of the latest Barclays report (heard it second hand). You can see the majors’ free cash flow (which kind of implies capex over time) here:

    It’s plain that the problem has not been (although is a bit now) a lack of high prices and implied demand. The problem is that these guys are being eaten alive on the cost side.

    Notwithstanding, the point stands. The levels of activity remain extraordinarily high. And productivity for conventional production is terrible. It’s quite hard to say just what the productivity of shale oil is. (John Kemp from Reuters took me to task for my agnosticism here:

    Personally, I think the productivity of oil production is a key driver of GDP for the country as a whole. But the problem, as Mark Sadowski has pointed out (and arguably John Fernald confirms) is that the effect is widespread. So you need a mechanism.

    In any event, it would be nice if economists prefaced their stagnation stuff with the simply extraordinary situation on the oil side. There is no lack of investment or activity on the oil side, just an inability to move the needle adequately–at least so far.

  23. Gravatar of Steven Kopits Steven Kopits
    26. September 2014 at 13:03

    Where is Sadowski, by the way? I don’t recall a comment from him in some time.

  24. Gravatar of Tom Brown Tom Brown
    26. September 2014 at 13:35

    “Where is Sadowski, by the way? I don’t recall a comment from him in some time.”

    In contrast to the rest of us, perhaps he’s found something valuable to do with his time. 😀

    Either that, or he has a girlfriend.

  25. Gravatar of Tom Brown Tom Brown
    26. September 2014 at 13:57

    … wait, I haven’t seen much of Morgan Wastler either. Do you suppose they had a duel, and did each other in?

    … or perhaps they crossed into a counter-factual universe, and can’t find a way back? Hmmm, who’s a good candidate around here [other than me] to send off into an alternate universe to go have a look see (possibly without ever returning!)? If only there was an obvious choice…

  26. Gravatar of ThomasH ThomasH
    26. September 2014 at 13:57

    Could it be that government have not responded optimally to low long tern rates by investing in projects with positive NPV’s at those rates?

  27. Gravatar of Tom Brown Tom Brown
    26. September 2014 at 14:03

    … and word is on the street, that most alternate universes are all about maximizing individual freedoms… filled to the brim with right-thinking individual(ist)s…. and the few that aren’t are full of people just dying to be lectured to at length about the subject. Probably just a rumor though…

  28. Gravatar of Lorenzo from Oz Lorenzo from Oz
    26. September 2014 at 14:29

    Scott: yes. But NGDP only a little.

  29. Gravatar of Various Various
    26. September 2014 at 22:24

    Well I think there are a few things going on here, and I’ll state right off that many of the following theories are probably wrong. I may also have some of my facts wrong. To the extent that any of the readers believe I have misstated any material fact, then by all means, please correct me. I have pondered this very question of why both real growth rates and inflation have been so low for so long, and consequently I am genuinely interested in some sort of unifying theory that explains it all. With that said, here are some thoughts:

    First, I think that part of the driver of lower real interest rates in the developed world has probably been the relative increase in the representation of developing country economies. Most of these countries (the most obvious and largest one being China) have usually run current account surpluses for about the past 20 years or so. As we know, countries that run current account surpluses tend to also run capital account deficits, which means that they must invest a lot of money abroad to make their balance of payments match. Most of these investments have flowed into assets of the developed world, including fixed income assets. Such demand for developed country assets has tended to push real yields down.

    Second, my understanding is that many of these developing countries, again with China being a good example, have purposely implemented various types of financial repression, with the objective of stimulating domestic investment at the cost of lowered consumer spending. This investment subsidy has lowered real domestic borrowing cost. Because capital controls are not perfect in most of these countries, the unusually low domestic borrowing costs have been partially exported from the developing countries to the developed countries. Here is one example of how this might happen. If there is an investor in China with a choice between investing in an artificially low yielding Chinese investment, or can choose instead to invest abroad, this investor will, on the margin, tend to invest abroad until the incremental demand for these developed country investments translate into a lower real yield. This marginal demand for developed country assets will cause their yield to, over time, decline until such yields are roughly equivalent with what can be obtained in China.

    Third, my understanding is that one of the objectives of financial repression in the developing countries is to cause a decoupling between asset price inflation and consumer price inflation. In other words, the financial repression is intended to have different effects on the people who are borrowing funds to invest in actual assets from the cohort of people that is buying consumer goods. I think it is quite possible that such decoupling was exported to the developed countries. As a result, you can have a situation where you experience significant asset price inflation, but little consumer price inflation. This gives the impression that real interest rates are lower than they would be under an alternative definition of “real”. Remember that the “real” interest rate is usually calculated as the nominal interest rate that a borrower pays to finance an asset purchase, less the rate of consumer price inflation. Thus the developed nation definition of “real” interest rates may suffer from an apples-and-oranges problem. The low real borrowing costs probably are causing significant inflation, except that this inflation is occurring in assets, and not in consumer goods. The high prices of assets may eventually translate into higher consumer prices, if the high asset prices are allowed to persist for a sufficiently long time because some sort of wealth effect and/or wealth transfer will occur between the investing and consumer cohort such that consumer price inflation will occur. But this can take a long time.

    In addition, it is possible that there are some important characteristics of both the developing and developed countries that are driving their particular capital flows, which in turn is causing both real growth and real interest rates in the developed world to be unusually low. With respect to the developing countries, my understanding is that the Asian Financial Crisis of 1997 – 1998 caused some developing sovereigns to conclude that having a relatively neutral current account balance may not be enough to stave off financial problems in the event of a “sudden stop” in currency flows. Specifically, my understanding is that some of these developing countries implemented policies to create larger current account surpluses than otherwise would have occurred. Because current account surpluses result in capital account deficits, this meant that these countries began to invest a lot of capital in developed countries. Second, it is possible that investors in these particular countries have, on average concluded that their domestic property rights could be in jeopardy, whereas developed countries are politically more stable, which means that such countries will not confiscate their investments. Consequently, these developing country investors may purposefully seek out investments in developed countries that have what appear on the surface to be inferior real yields.

    As a result of some combination of the aforementioned factors, real yields on developed country investments decline over time. At the beginning of such a cycle, this may cause the growth rate in the developed countries to be higher than normal, because asset prices rise and there is a wealth effect. Perhaps this is what happened in approximately the 1990s….I’m not sure. However, if investments in negative NPV projects continue unabated, the cumulative stock of overinvestment will eventually manifest itself in predominantly of low or negatively yielding real assets. Poor investment returns will cause the growth rate of the economy to decline because return on assets is a major driver of economic growth, or the lack thereof. In addition, the prospect of low real borrowing costs may cause the total debt burden in these economies to grow in a cumulative fashion, such that the rate of debt growth exceeds the gross rate of growth (i.e., the asset growth rate, as opposed to a leverage growth rate or growth rate of equity). As all finance types know, if the return on assets is less than the return on debt, the rate of increase in debt required to fund such investments is unsustainable in the long-run. Similar to how leverage tends to magnify returns to equity holders, in either a positive or negative manner, increasing levels of developed country sovereign debt is likely to magnify the economic cycles of the developed countries.

    So to summarize, I think you are right that a combination of both low interest rates and low growth seems like a conundrum. But I think this is because the sequencing of these events obscure the cause and effect relationship, and because it depends on whether you are talking about inflation of consumer prices or asset prices. I think the low real interest rate is probably more cause than effect, and that its origin is probably some combination of emerging economies exporting their artificially low interest rates and the developed countries appetite for cheap foreign capital to finance some of their negative NPV investments. I think the cause is more the former than the latter, but I’m not sure. However the slowing real growth rates of the developed countries is more effect than cause, and has been created by the cumulative stock of relatively poor NPV yielding investments in the developed countries, and made somewhat worse by the debt overhand that the developed sovereigns now operate under. In addition, asset price inflation and consumer price inflation in the developed countries have become decoupled when compared to historic norms, and as a result, the definitions of “real” rates of growth and borrowing have become somewhat muddied.

  30. Gravatar of
    27. September 2014 at 00:29

    I think you missed demography of the list of why bond yields are low across the board.

    As stylised facts young people have lower savings rate and require more capital (partly residential, but mostly it clearly right for companies to invest more in younger people as they benefit for longer). On the other hand, pensioners require no industrial capital at all.

    Thus a young population should have more investment and less propensity to save, and hence a higher ‘natural interest rate’ is required to match desired savings to desired investment. The US has much better demographics than , say Germany, which is much better again than, say Japan.

  31. Gravatar of blackswan blackswan
    27. September 2014 at 01:48

    Steven Koptits is correct. The world’s economy runs on oil, and oil has become too expensive to extract. And forget that falling US unemployment rate, and think, instead, about the falling labor participation rate. Worldwide, there is a severe labor glut in a period of growing de-industrialization.

    Energy has gotten as expensive as it can get, because people can’t and won’t pay higher prices for it. At the same time, it takes more energy to produce more energy. What we are seeing is not stagnation, but deflation.

  32. Gravatar of dtoh dtoh
    27. September 2014 at 03:13


    I’m not sure I’m following you. Nominal interests rates in each country simply reflect expected NGDP growth plus some risk premium. Since capital controls are quite limited (you can borrow in any currency), real expected growth in any one country (or to put in more correctly….currency) should not make that much of a difference to real rates (capital will just flow to where it can earn the best return). So in theory, the difference in nominal rates in different currencies should just reflect different inflation expectation…..which I think is consistent with the numbers.

    I guess I don’t see the mystery or maybe I’m missing something.

    BTW – There is no such thing as real interest rates except in hindsight.

  33. Gravatar of ssumner ssumner
    27. September 2014 at 07:50

    Patrick, That just makes the challenge all that much more fun!

    Matt, Given that employment is rising at 200,000/month, yes, I’d say 2% RGDP growth is really low. But of course slower labor force growth is part of the story, which reflect aging, and a lower LFPR also plays a role. Those are all supply-side factors.

    Todd, Airplanes were already very safe in 1970, although they’ve become even safer, as you say. But the big improvement was 1927 to 1970. Prices have fallen mostly due to deregulation, and also a bit due to better fuel efficiency. Progress in aviation isn’t even 10% as fast as in the old days.

    Brian, Yes, it’s partly demographics.

    Student, You are confusing pure theory, which makes no predictions about future growth, with theory plus assumptions, which does make predictions about future growth. But it’s a moot point—if those theories did imply fast growth then they are obviously wrong.

    Landing on Mars is much easier than putting men on the moon. We could have landed an unmanned probe on Mars in 1969, we simply chose to put men on the moon instead. There has been very little progress in space exploration technology since the invention of the Saturn 5 rocket in the 1960s.

    Steven, OK, but it’s a bit confusing when you base sweeping macroeconomic claims on capex in oil being low, and then 6 month later you base sweeping macro claims on capex being high. I can accept that it differs between majors and shale producers, but doesn’t that sort of cancel out, weakening your point?

    To be clear, I agree that the slowdown in productivity in oil is a part of the Great Stagnation story, but it’s a small part. Oil’s just not that big a share of the economy, and oil prices aren’t high enough to have a major impact.

    Thomas, That may be true, but it would not explain slow growth while unemployment is falling fast.

    Various, Asia is part of the interest rate story, but the China CA surplus is only 2% of GDP. And it doesn’t explain the slow growth in the US. If China invests in the US it should speed up US growth.

    Phil, Yes that’s one reason we are stagnating. I didn’t mention it because I thought it was obvious.

    Blackswan, A negative supply shock caused deflation?

    dtoh, The big mystery is why rates are so low, not why they differ between countries, which is a small mystery.

    I don’t understand your BTW–real rates are nominal minus expected inflation.

  34. Gravatar of Jason Smith Jason Smith
    27. September 2014 at 10:32

    Hi Scott,

    I know Canada has kept up it’s 2% target for awhile now — my comment/prediction was that it will fail to do so *in the future* (and maybe that is what markets are expecting in the 30 year interest rates causing the interest rate to fall below what would be expected if you assume 2% inflation over the next 30 years).

    That’s what makes Canada an interesting case (I’d say a test) for the information theory model: unlike other countries, it is close to the point where it should start to undershoot 2% inflation. The data from 2014 is inconclusive *either way*.

    Also, in the information theory model, employment is linked to RGDP growth (essentially Okun’s law), so inflation is not (to first order) relevant.

    I’ve written a bit more on the Great Stagnation here …

  35. Gravatar of Jason Smith Jason Smith
    27. September 2014 at 10:45

    Hi Scott,

    In your response to Student above you said: “Landing on Mars is much easier than putting men on the moon. We could have landed an unmanned probe on Mars in 1969, we simply chose to put men on the moon instead.”

    This is not actually true: the key technology is not just the size of the rocket (although that matters) but the guidance systems. The tolerance in your navigation is orders of magnitude tighter going to Mars relative to going to the moon.

    [This is also another reason why the space race got so much funding: landing on the moon is a comparable guidance problem to putting a nuclear weapon within 100s of feet of its target.]

  36. Gravatar of Todd Kreider Todd Kreider
    27. September 2014 at 12:38

    I agree with the above. I’d have to talk to a rocket scientist (do they still have those?), but I wouldn’t assume they could have landed a probe to Mars intact.

    Airline safety improved dramatically from the 1930s through 1970 but the graph I see (after Scott pointed it out) shows important gains from the 1970s to the present. Until the crash at SF that claimed two lives, there hadn’t been a passenger fatality since late 2001. (In a freak accident, a boy in a car died when an airliner ran off a runway.)

    Getting fatalities to almost zero from non-zero in the 1970s seems an important innovation for the psychological well-being of millions of passengers each year. They aren’t boarding a plane as they were in Chicago in 1980 remembering that a plane crashed on take off the year earlier, etc.

  37. Gravatar of dtoh dtoh
    27. September 2014 at 16:38

    I think I answered the small mystery. The big mystery you have answered a million times. Tight money = low expected NGDP growth. Not to mention the 60% increase in the capital gains tax.

  38. Gravatar of dtoh dtoh
    27. September 2014 at 19:53

    The more I think about this, the more I think the capital gains tax rate increase has had a big impact. If you assume any kind of asymmetry of returns on new businesses (a few winners and a lot of losers), it is possible (if not extremely likely) that the tax increase has pushed the average expected after tax return on new business investments to well below zero.

  39. Gravatar of Morgan Warstler Morgan Warstler
    27. September 2014 at 21:57

    It’s terribly unhealthy for old men economists to not confront this:

    We have flying cars right now. You put on your VR rig, and you not only fly in a car, but on a the back of a dragon, while you chase down and capture members of the Swedish bikini team. You can be you, the car, the dragon, the bikini team or the bikini.

    It’s hard for old men to admit that consumers have been CHOOSING a move to freely copyable digital entertainment. It isn’t “the consolation prize” for no flying car, it is the BEST PRIZE AT THE SKEE BALL TICKET COUNTER.

    What are drones? Drones = you stay at home, plugged into the network, behind your security system, and have your atoms delivered to you.

    Mop your floor? Scott, exactly how many Roombas have your purchased?

    Look, Scott, before you type a bunch of words into this box which is the only reason anyone has ever heard of MM or Tyler…

    more next so I can use another link….

  40. Gravatar of Morgan Warstler Morgan Warstler
    27. September 2014 at 22:07

    Scott, before you go on and on again, first you need to go to this:

    If 200 years from now, we do not have Scott’s atomic flying cars, I will not be surprised.

    If 200 years from now, there isn’t a software program that continue to let approximations of me and Scott argue about why there are no flying cars, I wil be surprised.

    NONE OF THE ABOVE FROM SCOTT deals with this: GDP that grows at 1% per year, CAN deliver 5% invisible RDGP that delivers 4% of deflation.

    If there is such a thing as invisible deflation.

    In the face of seeming contradictions, science shouldn’t have a problem hypothesizing an unobservable thing exists.

    Invisible Real Growth in a world where everything anybody wants is copyable for free, in a world where normal people CLEARLY do not want atomic inventions that old men long for….

    Hypothesizing unobservable things is the scientific answer.

  41. Gravatar of ssumner ssumner
    28. September 2014 at 05:41

    Jason and Todd, I don’t agree about the landers. Lots of US and Russian probes were sent to Mars in the 1970s. Some failed and some succeeded. Russia put a probe on Venus in 1982. There has been a bit of progress since (obviously the computer guidance is better), but nothing dramatic.

    Todd, People SHOULD feel safer flying, but they don’t. So the psychic gain is not there, and the real gains from the 1970s are small, as flying was already pretty safe in the 1970s. But I already admitted it is extremely safe today.

    dtoh, The cap gains rate explains very little. It’s had been falling for 30 years, until last year, just as long term interest rates had been falling for 30 years. The mystery is low long term real rates, which cannot be explained by monetary policy. So I don’t think you do agree with me on the big mystery.

    Morgan, I thought Roombas vacuumed, not mopped.

    You may be right about progress, but free stuff doesn’t cause higher real interest rates. Investors only care about investment with a monetary return. That’s what has fallen.

  42. Gravatar of Various Various
    28. September 2014 at 20:21

    Scott, I think China’s current account surplus has averaged about 4% of GDP during the past 20 years, not 2%. However, that number has bounced around a lot during the 20 years. Here’s one graph:

    I think China’s investment in the U.S. has created asset bubbles. I don’t see why these investments would necessarily create GDP growth, especially if the asset bubbles have a habit of popping, which I think would be disruptive. I also don’t think such investment will be growth positive if the assets it invests in produce poor returns. BTW, I’m on board with the secular stagnation hypothesis.

  43. Gravatar of ssumner ssumner
    29. September 2014 at 05:24

    Various, I was discussing the current (and future) growth stagnation; 20 years ago we were booming.

    The standard economic model says more investment raises growth. You are going to need a much more persuasive argument to displace that presumption.

  44. Gravatar of Danilushka Danilushka
    29. September 2014 at 05:44

    Because the demand-side in a very consumer-driven nation (guess which one) is sputtering due in increasing concentration of wealth which will eventually lead to deflation 1930s-style. The supply side won’t invest much (in infrastructure or people) they see demand rise so the dog chases it’s tail.

  45. Gravatar of dtoh dtoh
    29. September 2014 at 09:33

    Are you saying real interest rates are falling relative to RGDP growth rates. Or are you saying the long term trend for both is lower rates.

    As for the cap gains rate, I was referring to your statement;

    “This is the only recovery I ever recall seeing where the growth rate of real GDP in the US was below the trend growth rate.”

Leave a Reply