The Future of Global Interest Rates — Trending Lower?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

The Future of Global Interest Rates — Trending Lower?

Post by SimpleGift »

This post examines the relationship between a country's interest rates and the age of its population — and finds specifically that the younger a country's population, the higher its "natural" rate of interest, and the older a country's population, the lower its "natural" interest rate. Extending this insight, what does it suggest for the future of global interest rates in a steadily aging world?

In the first chart below, the current 10-year government bond yields for 25 countries are plotted against the median age of each country's population today. The correlation is eye-popping (R^2 = 0.88). But why? The best explanation I've found is that young people are generally borrowers, creating the demand for capital, and middle age and older people are generally savers, creating the supply of capital. When there are more younger people than older in a population (think India, Mexico or Indonesia), demand for capital is stronger than supply and interest rates are higher. When older people outnumber younger people (think Japan or Germany), supply outpaces demand and interest rates are lower.

Image
10-year Government Bond Yields as of 2/11/15, from Trading Economics
Country Median Age Data for 2015, from United Nations Population Division

DISCUSSION: We're all aware that central banks around the world peg the short end of the yield curve and primarily determine short term interest rates. But intermediate and longer term interest rates are more determined by the supply and demand for capital. In an aging world, where the supply of savings is steadily growing due to more older folks, and the demand for capital is steadily falling due to fewer younger folks, it's hard for me to see how interest rates are going to return to historical levels anytime soon. Your thoughts?

Image
World Median Age Data from United Nations Population Division
Last edited by SimpleGift on Wed Feb 11, 2015 7:22 am, edited 1 time in total.
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

It is a nice chart with a good story to go with it. I have a few criticisms:

1. Why these 25 countries? I believe you have dropped some outliers.

2. Differences in nominal interest rates across countries are largely a function of expected inflation. The argument would be stronger if you used real rates, as this reflects demand for capital (you might want to argue demographics explain inflation, but that's a different story). While not all of these countries issue inflation-linked bonds, I suspect the correlation with demographics is much weaker among those countries where information on real rates is available.

3. Most of the correlation appears to come from comparing a few "top-left" EM countries (India, Indonesia, Mexico, etc.) with "bottom-right" more developed markets. As in point 2, I believe inflation is providing much of the difference between these two cohorts. If you compare developed markets within-cohort, the correlation mostly goes away.

4. It isn't really fair to use different dots for each EU country, as they all have the same real rate (credit-quality adjusted). In other words, differences between German and Italian rates do not represent differences in the savings propensities of different demographics, but rather how the market evaluates creditworthiness--how much more likely you are to get your money back from lending to Germany vs. Italy.

5. If you looked at this graph in earlier time periods, it would not show such a strong cross-sectional correlation. So why should we expect the correlation to last going forward?
richard
Posts: 7961
Joined: Tue Feb 20, 2007 2:38 pm
Contact:

Re: The Future of Global Interest Rates — Trending Lower?

Post by richard »

market timer wrote:2. Differences in nominal interest rates across countries are largely a function of expected inflation. The argument would be stronger if you used real rates, as this reflects demand for capital (you might want to argue demographics explain inflation, but that's a different story). While not all of these countries issue inflation-linked bonds, I suspect the correlation with demographics is much weaker among those countries where information on real rates is available.
Do you have evidence differences are a function of expected inflation? How are you measuring expected inflation for countries without inflation-linked bonds?
market timer wrote:3. Most of the correlation appears to come from comparing a few "top-left" EM countries (India, Indonesia, Mexico, etc.) with "bottom-right" more developed markets.<>
This was my first thought - the upper left EM countries are strongly influencing the result. EM v. developed markets seems a very plausible explanation for interest rates.

My impression is EM countries tend to grow faster than developed markets and there tends to be a relation between faster growth and higher real rates.
Valuethinker
Posts: 48944
Joined: Fri May 11, 2007 11:07 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by Valuethinker »

The R2 is so high it makes me think the dependent and independent variables are somehow correlated.

Capital markets are not perfect. Underdeveloped countries don't have fully functioning domestic capital markets. Since consumers and small businesses can't borrow (except from moneylenders often at usurious rates of interest) people have very high savings rates. Also there are limited or no state pension schemes, so your children are your retirement pension.

It's a good bet that interest rates will be low in the future. Partly due to global demographics but also due to the apparent demand shortfalls-- the Secular Stagnation thesis. Certainly Japan hasn't managed to raise interest rates, Europe looks like the next Japan, China is having a financial slump. I suspect it's a long time before we see 'normal' interest rates of say 5% again.
User avatar
Topic Author
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by SimpleGift »

market timer wrote:Why these 25 countries? I believe you have dropped some outliers.
Appreciate your thoughtful critiques, market timer. I tried to make sure that the major developed and emerging economies were represented. That's possibly why the chart has a "barbell" appearance. The only outlier that I consciously excluded was Russia, since its interest rates today are clearly distorted far beyond their "natural" range.

Here's the complete list of 25 countries (from youngest median age to oldest): South Africa, India, Mexico, Indonesia, Turkey, China, China, Australia, United States, Thailand, Norway, United Kingdom, Canada, Korea , France, Sweden, Belgium, Spain, Switzerland, Netherlands, Hong Kong, Austria, Italy, Germany and Japan.
market timer wrote:If you looked at this graph in earlier time periods, it would not show such a strong cross-sectional correlation. So why should we expect the correlation to last going forward?
It would be interesting to compare earlier time periods to see if there's still a rough correlation between median age of population and interest rates. However, I feel the under-appreciated point is that interest rates (at least intermediate to longer term rates) are mainly a function of supply and demand for capital — and I'd expect this relationship to hold up globally going forward. Especially in the developed economies, it appears there will be a strong demographic headwind toward lower interest rates for decades to come.
garlandwhizzer
Posts: 3562
Joined: Fri Aug 06, 2010 3:42 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by garlandwhizzer »

Another great post, Todd. I'd like to offer another point to consider in explaining why bond yields get lower as median age increases. As we get older our portfolios tend more and more to slant toward fixed income for safety and yield. Also older average populations tend to be in more affluent (non-emerging) countries, and therefore having more money to buy these bonds. Since their spending is in their local currency they would tend to buy their bonds from their own countries, accepting lower yields if necessary in return for safety, familiarity, and local currency. In short the natural risk aversion that comes with age plus the increased assets that come with developed markets might act together to lower bond yields in developed countries with aging populations as opposed to emerging countries with younger populations.

Garland Whizzer
User avatar
Topic Author
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by SimpleGift »

garlandwhizzer wrote:In short the natural risk aversion that comes with age plus the increased assets that come with developed markets might act together to lower bond yields in developed countries with aging populations as opposed to emerging countries with younger populations.
Good point, GW. No doubt the aging populations in the developed countries today are already increasing the global demand for bonds and other yielding assets — and this trend will inexorably continue in the decades ahead (chart below). These demand pressures could well contribute to keeping yields low throughout our lifetimes.

And soon it won't just be the developed countries that are aging and wanting to buy more bonds — the emerging countries, as a whole, are also beginning to age and will be demanding more safe assets as well. According to the UN, in 30 years the emerging countries in aggregate will have nearly 1 billion more people over the age of 60 than they have today!

Image
Demographic Data from United Nations Population Division
stlutz
Posts: 5585
Joined: Fri Jan 02, 2009 12:08 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by stlutz »

In an aging world, where the supply of savings is steadily growing due to more older folks, and the demand for capital is steadily falling due to fewer younger folks, it's hard for me to see how interest rates are going to return to historical levels anytime soon. Your thoughts?
I don't see it this way.

Older folks have negative savings rates--that is, they have money they saved up and then they are just consuming it. Such a reduction in savings would cause the cost of borrowing to increase, not decrease.

You are correct that they also borrow less, which would cause lower rates.

At least in some countries, their versions of Social Security are funded in part by borrowing, which causes higher rates.

In short, I don't see any logical relationship between demographics and real rates.

Market timer may be onto something by pointing in the direction of a connection between demographics and inflation. I think the argument here is that when you have a lot of labor, there is no incentive to improve productivity. When there are not productivity gains, then higher wages simply lead to higher prices. For example, the growth rate of the labor force in the US peaked in the late 70s.
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

richard wrote:
market timer wrote:2. Differences in nominal interest rates across countries are largely a function of expected inflation. The argument would be stronger if you used real rates, as this reflects demand for capital (you might want to argue demographics explain inflation, but that's a different story). While not all of these countries issue inflation-linked bonds, I suspect the correlation with demographics is much weaker among those countries where information on real rates is available.
Do you have evidence differences are a function of expected inflation? How are you measuring expected inflation for countries without inflation-linked bonds?
I'm just using the following identity: Nominal rate = real rate + expected inflation. For countries that don't issue inflation-linked bonds, backing out an ex ante real rate is not really possible. However, clearly, those "top-left" countries have much higher expected inflation than the "bottom-right." See, for example, the differences in realized inflation over the past year here: http://www.global-rates.com/economic-in ... ation.aspx

Inflation running at 6% in India and 7% in Indonesia substantially erodes the buying power of the 7-8% 10-year bonds. I don't know what their inflation expectations are for the next 10 years, but they're surely higher than in the US, Eurozone, and Japan. This supports the EM vs. DM inflation story, not the cost of capital story.
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

Simplegift wrote:It would be interesting to compare earlier time periods to see if there's still a rough correlation between median age of population and interest rates.
Agreed, and I do think you highlight an interesting relationship between yields and demographics. I just don't believe the relationship is as clear as your graph suggests.
Angst
Posts: 2968
Joined: Sat Jun 09, 2007 11:31 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by Angst »

stlutz wrote:
In an aging world, where the supply of savings is steadily growing due to more older folks, and the demand for capital is steadily falling due to fewer younger folks, it's hard for me to see how interest rates are going to return to historical levels anytime soon. Your thoughts?
I don't see it this way.

Older folks have negative savings rates--that is, they have money they saved up and then they are just consuming it. Such a reduction in savings would cause the cost of borrowing to increase, not decrease.
"Older folks" don't necessarily have negative savings rates. Consider the "age in bonds" notion that so many people seem to approximate if not literally follow. The older you get, the more you move out of equity and into bonds. Yes, at some point you do have a negative savings rate, but does that overwhelm your ongoing movement from equity to bonds, let alone the people younger than you who are also aging and slowly moving from equity to bonds?
User avatar
Topic Author
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by SimpleGift »

market timer wrote:Inflation running at 6% in India and 7% in Indonesia substantially erodes the buying power of the 7-8% 10-year bonds. I don't know what their inflation expectations are for the next 10 years, but they're surely higher than in the US, Eurozone, and Japan. This supports the EM vs. DM inflation story, not the cost of capital story.
Your point is well taken that inflation is a big part of the observed difference in nominal rates between emerging and developed countries today. Thanks for clarifying this. But I'd also argue that the very same demographic trends that are creating slow growth and low real interest rates in the advanced economies today (due to a lack of demand for capital) are also exerting downward pressure on inflation. Japan and Europe, with the most rapidly aging populations, are experiencing the lowest inflation and even deflation (chart below).

In other words, steadily aging populations today (especially in the developed economies) are putting downward pressure on real interest rates as well as inflation and thus nominal rates — and this trend toward lower yields is likely to continue for decades to come. The global demographic winds all appear to be blowing in the same direction from here forward.

Changes in Age Dependency Ratios and Average Inflation (%), 2009-2014
Image
(NOTE: The age dependency ratio is the ratio of dependents (people younger than 15 and older than 64) to the working-age population (those aged 15-64). Changes in both dependency ratios and inflation rates are 5-year averages, 2009-2014.)
Source: Blackrock
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

Simplegift wrote:Your point is well taken that inflation is a big part of the observed difference in nominal rates between emerging and developed countries today. Thanks for clarifying this. But I'd also argue that the very same demographic trends that are creating slow growth and low real interest rates in the advanced economies today (due to a lack of demand for capital) are also exerting downward pressure on inflation. Japan and Europe, with the most rapidly aging populations, are experiencing the lowest inflation and even deflation (chart below).
It still needs to be established that demographics are pulling down real interest rates to any meaningful extent. I'm not sure that there is much of a cross-sectional correlation. Part of the reason for this is how easily capital can move between countries. In fact, to attract foreign capital, many of these countries issue debt denominated in a currency other than their own.
In other words, steadily aging populations today (especially in the developed economies) are putting downward pressure on real interest rates as well as inflation and thus nominal rates — and this trend toward lower yields is likely to continue for decades to come. The global demographic winds all appear to be blowing in the same direction from here forward.
I'm less confident in this long run prognosis. While I agree with you that growth will slow--in part from demographic factors, in part from slower credit growth--I'm not sure how much room there is for long term real rates to move lower. Given the magnitude of central bank intervention since the GFC, I don't believe we are looking at truly two-sided markets in these yields, especially Japan. In equilibrium, these long term bonds must compete with positive real return investments and zero real return commodities for capital, so I think we are near a natural lower bound for real rates. As for nominal yields, these are largely a function of inflation, and inflation in the long run is explained by growth of money supply, not demographics. Central banks are fickle and monetary regimes change every generation or two. It's almost like a game of three card monte--just when you think you know what to expect, your wealth is revealed to have been an illusion. The current regime, since the chaos that followed the breakdown of Bretton Woods, is based on ever declining nominal rates, supporting ever greater leverage and asset price inflation, but now we are reaching the limits of this--the ZLB, debt saturation, and secular stagnation are all symptoms. Once we reach the end of the current regime, I expect the rules will have an abrupt change, leaving many who expected low inflation disappointed.
Valuethinker
Posts: 48944
Joined: Fri May 11, 2007 11:07 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by Valuethinker »

market timer wrote: I'm less confident in this long run prognosis. While I agree with you that growth will slow--in part from demographic factors, in part from slower credit growth--I'm not sure how much room there is for long term real rates to move lower. Given the magnitude of central bank intervention since the GFC, I don't believe we are looking at truly two-sided markets in these yields, especially Japan. In equilibrium, these long term bonds must compete with positive real return investments and zero real return commodities for capital, so I think we are near a natural lower bound for real rates. As for nominal yields, these are largely a function of inflation, and inflation in the long run is explained by growth of money supply, not demographics. Central banks are fickle and monetary regimes change every generation or two. It's almost like a game of three card monte--just when you think you know what to expect, your wealth is revealed to have been an illusion. The current regime, since the chaos that followed the breakdown of Bretton Woods, is based on ever declining nominal rates, supporting ever greater leverage and asset price inflation, but now we are reaching the limits of this--the ZLB, debt saturation, and secular stagnation are all symptoms. Once we reach the end of the current regime, I expect the rules will have an abrupt change, leaving many who expected low inflation disappointed.
I think the problem with your model is that Central Banks don't choose inflation rates? Particularly not in an era of global capital mobility.

Just as the Central Banks in the last 60s and 70s were 'behind the curve' on inflation, so they are now behind the curve on the deflationary forces that are out there.

In other words, this is a real story, not a nominal one? There are real economy forces out there which mean an inflation of the money supply doesn't lead to an increase in lending, economic activity and inflation? The world Japan has gotten too, we are now all tending towards.

Since monetary policy in part works by the devaluation channel, and we can't all devalue our currencies against each other, that's one reason why you don't get general inflation.

You'd have to look for an end to those real forces, amongst which we might include:

- technology and particularly the rise of robots and robot agents - self driving cars could replace 10s of millions of minicab drivers

Looking at the history of electric motors and factories (it was literally decades before EM was universal in factories) it's likely that the internet and the subsequent mobile revolution have only begun to alter business practices

- decline in reserve banking as the channel for moving money around the economy: you have the shadow banking system and the bond markets, you have peer to peer lending

-demographics - the planet as a whole is aging rapidly *except* for a group of countries in the Middle East and Africa. If businesses start relocating labour intensive operations to those countries, you get another leg downward in the trend that began with China in 1979 (modernization, opening up to the world economy)

- even in something as mature as transport you are still seeing innovation in air transport (more efficient planes think Dreamliner); shipping (wider Panama canal etc.); railroads (new efficiencies there)

Countervailing that you might see rising resource prices (richer people consume more). That was certainly the story with the peak of world grain prices around 1910 (all time in recorded history).
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

Valuethinker wrote:I think the problem with your model is that Central Banks don't choose inflation rates?

In other words, this is a real story, not a nominal one? There are real economy forces out there which mean an inflation of the money supply doesn't lead to an increase in lending, economic activity and inflation? The world Japan has gotten too, we are now all tending towards.
I do believe central banks can choose inflation rates. As a Neo-Fisherian, I think the (nominal) policy rate determines the rate of inflation in equilibrium--that is, higher interest rates lead to higher inflation in equilibrium. In the past 30 years, central banks could rely on households and businesses borrowing more in response to lower interest rates (e.g., you can qualify for a larger mortgage for a given income as mortgage rates decline). This created money via debt, and, in the short run, offset the tendency for lower rates to lead to lower inflation. Now we've reached the limit at the ZLB, and the way to get higher inflation is actually to start paying higher interest. Of course, there may be some short term pain as borrowing declines and asset prices fall. Japan has only demonstrated to me the futility of the status quo, and its high debt means its transition to a higher rate environment will be particularly difficult. This is the future I think we are headed toward: higher nominal rates, higher inflation, lower growth, lower asset prices.
richard
Posts: 7961
Joined: Tue Feb 20, 2007 2:38 pm
Contact:

Re: The Future of Global Interest Rates — Trending Lower?

Post by richard »

market timer wrote:
richard wrote:
market timer wrote:2. Differences in nominal interest rates across countries are largely a function of expected inflation. The argument would be stronger if you used real rates, as this reflects demand for capital (you might want to argue demographics explain inflation, but that's a different story). While not all of these countries issue inflation-linked bonds, I suspect the correlation with demographics is much weaker among those countries where information on real rates is available.
Do you have evidence differences are a function of expected inflation? How are you measuring expected inflation for countries without inflation-linked bonds?
I'm just using the following identity: Nominal rate = real rate + expected inflation. For countries that don't issue inflation-linked bonds, backing out an ex ante real rate is not really possible. However, clearly, those "top-left" countries have much higher expected inflation than the "bottom-right." See, for example, the differences in realized inflation over the past year here: http://www.global-rates.com/economic-in ... ation.aspx

Inflation running at 6% in India and 7% in Indonesia substantially erodes the buying power of the 7-8% 10-year bonds. I don't know what their inflation expectations are for the next 10 years, but they're surely higher than in the US, Eurozone, and Japan. This supports the EM vs. DM inflation story, not the cost of capital story.
I agree that an EM v DM story is the more plausible explanation for the chart in the OP. However, there does not appear to be any way to determine the extent to which higher rates are due to expected inflation rather than higher real rates. EM economies tend to have faster real growth than DM economies, which would support the idea that real rates are higher. They also have higher inflation, which suggests higher expected inflation.

It's likely a combination of the real rates and expected inflation.
User avatar
Topic Author
SimpleGift
Posts: 4477
Joined: Tue Feb 08, 2011 2:45 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by SimpleGift »

market timer wrote:It still needs to be established that demographics are pulling down real interest rates to any meaningful extent. I'm not sure that there is much of a cross-sectional correlation. Part of the reason for this is how easily capital can move between countries. In fact, to attract foreign capital, many of these countries issue debt denominated in a currency other than their own.
I'd like to run the same regression as in the OP, but using real rates instead (i.e, a country's median age vs. its 10-year inflation-linked bond yield). Apparently quite a few developed countries now issue inflation-linked bonds, as do 8 or 9 emerging countries. However, I've not been able to find data on current yields for enough countries to do the comparison (no Bloomberg terminal at my desk!). If anyone can provide a source (or the data itself) for current yields on 10-year inflation-linked bonds from, say, 20-25 countries (both developed and emerging), I'll be happy to re-run the regression. Thanks!

My suspicion is that market timer is right — the correlations will be weaker, due to the cross-border movement of capital — but I expect there will still be some positive correlation, reflecting the markedly different demographics today between the emerging and developed countries.
richard
Posts: 7961
Joined: Tue Feb 20, 2007 2:38 pm
Contact:

Re: The Future of Global Interest Rates — Trending Lower?

Post by richard »

market timer wrote:<>I do believe central banks can choose inflation rates. As a Neo-Fisherian<>
A bit off-topic so I'll just link to a post "let’s blow the neo-Fisherite model right out of the water" http://www.themoneyillusion.com/?p=26671
User avatar
market timer
Posts: 6535
Joined: Tue Aug 21, 2007 1:42 am

Re: The Future of Global Interest Rates — Trending Lower?

Post by market timer »

richard wrote:
market timer wrote:<>I do believe central banks can choose inflation rates. As a Neo-Fisherian<>
A bit off-topic so I'll just link to a post "let’s blow the neo-Fisherite model right out of the water" http://www.themoneyillusion.com/?p=26671
As it's off-topic, I'll just post this recent article, showing how it's a different ballgame with the Fed paying interest on reserves: http://faculty.chicagobooth.edu/john.co ... policy.pdf
lee1026
Posts: 396
Joined: Sat Jun 21, 2014 7:47 pm

Re: The Future of Global Interest Rates — Trending Lower?

Post by lee1026 »

In equilibrium, these long term bonds must compete with positive real return investments and zero real return commodities for capital, so I think we are near a natural lower bound for real rates.
Commodities do have risk as well as storage costs. It is not obvious that 0% +- 10% is better than -2% guarenteed.
Post Reply