Did the "Savings Glut" Just Go Into Reverse?
Shortly before he handed the mess he had crated to Ben Bernanke (who is now busy creating an even bigger one), Alan Greenspan muttered something about a "conundrum" with regard to long term Treasury bond yields. The conundrum at the time was 'why does the bond market not sell off in view of the Fed hiking short term rates'?
The so-called conundrum created an excellent excuse allowing the Fed to deny its responsibility for the housing bubble. See, the bubble wasn't the Fed's fault, since the Fed was evidently not responsible for long term rates staying low. The culprit was a so-called "savings glut", created by evil savers in Asia. Now, it is true that a great many Asians are saving a lot – they always have. It's a mentality thing. In fact, the global savings rate was higher, on average, in the 15 years preceding the housing bubble than during the housing bubble (although, as you will see below, the "savings rate" as it is defined today is a dubious concept). So how can all those savings have "caused" a housing bubble in one instance but not the other? As Frank Shostak pointed out in 2005, Greenspan and Bernanke's problem was that they confused "money" with "savings":
“Since mainstream thinking views an excess of saving, or too much of it, as bad news for economic activity, obviously what Greenspan also had in mind in his speech was that world economies may be under threat due to a glut of savings. Indeed many economists are of the view that too much saving could destabilize the world economy.
However, does all this make much sense? It seems that most experts including Greenspan have fallen into the trap of confusing money with saving.
The notion of the supposed world glut of savings is based on the premise that saving is the amount of money left after monetary income was used for consumer outlays, which implies that saving is synonymous with money. For a given amount of monetary expenditure on consumer goods an increase in money income implies more monetary saving.
However, what matters for economic growth is not monetary saving but rather the stock of real savings. This stock, however, cannot be quantitatively ascertained because of the heterogeneous nature of final goods and services. We cannot arithmetically add up potatoes and bread into a meaningful total. The employment of various price deflators to extract out of monetary income real income and in turn real saving will not do the trick since it contradicts the fact that potatoes and tomatoes can’t be added up to a meaningful total. One thing we can be assured is that monetary pumping can never be good for the pool of real savings.
Most so-called savings countries have actually been engaged in strong monetary pumping over the past six years. So it is quite likely that on account of loose monetary policies the strong monetary saving in fact is not that strong in real terms. For instance, between January 99 to June 2005 China’s money M1 increased by 153%. Malaysian money M1 increased during this period by 107%, Thailand’s money M1 rose by 93% while money M0 in Russia increased by 828%.
To be sure, qualitatively one can suggest that countries like China and the former Soviet Union are generating more real wealth than in previous times on account of the introduction of a freer market economy. This, we suggest, has given an important support to the US economy. Because the US dollar is an internationally accepted medium of exchange, through monetary expansion Americans can divert real savings from other countries, i.e., they can engage in an exchange of nothing for something. This ability to divert world real savings to the US doesn’t mean that there is abundance of real savings in the world.”
Yep, enormous monetary inflation and "savings" are not the same thing. Who would have thought? Not Greenspan and Bernanke, obviously. Moreover, monetary inflation in the U.S. is actually under the Fed's purview. China cannot increase U.S. asset prices simply by printing more yuan – someone who possesses yuan has to first buy dollars before he can embark on a spending spree in the U.S. And the supply of dollars is entirely "home made", by U.S. commercial banks in their lending operations and the Fed which accommodates them – these are the entities that create U.S. dollars from thin air, which then in turn can be used to push up asset prices in the U.S.
In addition, as George Reisman has pointed out, it is simply absurd to claim that China, Venezuela, Mexico, Russia, Nigeria, Thailand, and so forth can possibly have a 'glut of savings' when the capital intensity of their economies is a fraction of that of the developed world. And assuming that their capital intensity were to increase to the point where it is equal to that of the developed world, it would still not be possible for a "glut" of savings to emerge in the market, on account of the fact that the scarcity of capital is not going to disappear. Moreover, if savings increase, then both more capital goods and more consumer goods will ultimately be produced, which in the absence of an increase in the money supply will inevitably lead to a decline in their prices. It is completely bizarre to ascribe a bubble in asset prices to a "savings glut". What creates bubbles is the increase in the credit and money supply. Aren't you comforted to learn that both the current Fed chairman as well as his predecessor apparently were and remain blissfully unaware of such pesky facts?
So obviously we posed a trick question above: interest rates have not been rising lately because the alleged "savings glut" has suddenly reversed, because there never was and there still isn't a savings glut in existence.
Why Have Bond Yields Risen?
It is widely claimed that bond yields are suddenly defying the Federal Reserve's massive debt monetization program because the market "sees better times for the economy ahead". The argument basically goes: if the economy gets better, the Fed will tighten monetary policy, and since the market discounts this better future, yields are rising in advance.
Funny, that. We are slightly mystified what has happened to the savings glut. Why has it suddenly and quietly disappeared from the argument? We're not aware of the savings habits of emerging market denizens having somehow changed all of a sudden (consider that in three lectures between 2005 to 2007, Bernanke used the term "savings glut" altogether 30 times, so to see it disappear into the memory hole since then is certainly odd).
Obviously, something else must be going on. Most observers have probably noticed that Treasury bond yields have since 2008 tended to decline ahead of the Fed's "QE" programs, and tended to rise during their implementation. This alone proves that even though it is a big force in the market, the Fed can actually not influence long term interest rates by buying a fixed amount of Treasuries every month. It could in theory set a ceiling on rates by committing to buying as many bonds as it takes if yields were to rise above the ceiling value, but it should be obvious that such a scheme could very easily blow up.
Anyway, we have to look for another explanation for rising yields during 'QE' operations, and in order to explain them one must consider what market interest rates of bonds consist of. In brief, there is the natural interest rate – the discount people place on future versus present goods, a risk premium that relates to the risk of the borrower and a price premium (or inflation premium), that relates to the expected future devaluation of money. A major feature of previous iterations of "QE" were rising inflation expectations (as measured by comparing either the prices or yields of "inflation protected" to those of "normal" bonds) – this is to say, expectations that CPI would rise faster in the future (this happens irrespective of the fact that CPI is a flawed datum).
Unfortunately, this correlation has recently broken down, and that is the reason why there is now a "conundrum". 10 year note yields have done this:
The 10-year Treasury note yield has been rising rather forcefully of late.
Concurrently, 10-year inflation break-evens (the 10 year nominal constant maturity yield minus the 10-year inflation-indexed securities constant maturity yield) have done this:
10 year inflation break-evens plunge to a 17 month low.
The gap between yields on 10-year notes and Treasury Inflation-Protected securities of comparable maturity narrowed to as low as 1.91 percentage points, a level unseen since Jan. 3, 2012. The so-called breakeven rate represents the bond market’s expectations for the rate of growth in consumer prices during the life of the debt.
U.S. government data may show on June 27 that the core personal-consumption-expenditure price index probably rose 1.1 percent in May from a year earlier, according to the median forecast of economists surveyed by Bloomberg. The Fed’s preferred measure of inflation slid to 1.05 percent in April, the lowest on record dating back to 1960.”
Actually it appears that this recent move in inflation expectations correlates much better with gold than with Treasury bond yields – so what gives? It seems unlikely that yields are rising because traders are suddenly worrying about the government's solvency after all (they may one day, but we don't think that day has come yet).
Did Japan Provide the Trigger?
We can only supply a guess at this stage. We have noticed that another data series seems to have led the recent decline in inflation expectations – namely the Japanese yen. Take a look:
The Japanese yen, weekly.
The yen's decline began in earnest when it became clear that Shinzo Abe would likely win the election. Once he did win, the decline accelerated and became relentless. Another shot in the arm to the swoon was provided by the implementation of the BoJ's latest crazy inflationary scheme under its new governor Kuroda.
So why is there a connection? We think there are two aspects that are important in this context. One is that due to the yen's sharp decline, goods exported from Japan - which is after all one of the world's largest exporters - are becoming cheaper outside of Japan. This gives an incentive to its most prominent competitors to also join the "currency war", as they fear they will otherwise lose out. Consider e.g. that the Korean Won has also begun to decline since the beginning of the year, and is now getting close to lateral support levels:
The Korean Won has begun to decline as well.
So bond traders may have gotten the idea that consumer goods prices will come under pressure as the effects of Japan's yen devaluation spread; in addition, economic data are not very strong, with Europe in recession and recent U.S. data not much to write home about either – hence falling inflation expectations (remember, these expectations only concern CPI, they are neither indicative of the pace of monetary inflation nor of the likely actual future loss of money's purchasing power).
But why are Treasury bond yields rising while inflation expectations are declining? Again, we believe Japan's central bank may actually be the culprit. One (unstated) intention of its pumping maneuver was to get Japanese investors to hop aboard the 'carry trade' – selling yen in order to purchase higher yielding assets abroad. For one thing though, Japanese institutions did not react as expected – instead they reportedly have been selling overseas assets to book the unexpected windfall stemming from exchange rate gains.
For another thing, the BoJ's declared inflation goal cannot coexist with JGB yields at 40 or 50 basis points. The resulting increase in the volatility of JGB yields has hit many players in that market harder than it may appear from the actual increase in yields as such. Now consider that bond markets are often played by employing a lot of leverage. This magnifies the proceeds from yields, and the temptation to increase one's leverage is especially great when yields are slowly declining, central banks are promising to keep them low and there is little volatility.
However, the market reaction to the BoJ's interventions changed all that, and just as if someone had thrown a rock into a pond, the effects have rippled outward from Japan. Japanese investors are among the biggest in the world after all, so when they started selling assets to take profits (or perhaps to shore up their cash reserves in view of rising JGB volatility), that very likely created problems for other leveraged players, whose selling in turn hit the next group and so forth - as evidenced by the recent massacre in EM bonds and currencies as well as municipal debt, which had all the hallmarks of forced selling (China's slowing growth has further increased EM woes of course).
There is no reason to assume that holders of Treasury bonds don't use a lot of leverage. In fact, due to their "risk-free" status, margin requirements for Treasuries are tiny. A rise in Treasury yields in turn tends to be exacerbated by "convexity selling", i.e. the attempt to bring the duration of large mortgage backed bond portfolios into line. As previously discussed, hedging of portfolios of mortgage bonds against either prepayments when yields are declining or in order to maintain duration and hedge against capital losses when yields are rising, invites buying and selling in a highly pro-cyclical manner. In short, there is a lot of trading going on that is practically on autopilot and tends to lead to exaggerated moves in both directions.
However, the trigger for all this commotion was very likely Japan, or rather, the policies pursued by Shinzo Abe and Haruhiko Kuroda's BoJ. It seems likely that they are responsible for both the recent sharp decline in inflation expectations and the rise in bond yields around the world – it is a bit reminiscent of chaos theory, even if the BoJ is more like an angry elephant storming into a China shop than a butterfly flapping its wings. What this shows is that there is way too much leverage in the system, something that is a direct result of employing a fiat money that can be expanded at will, and has in fact been expanded at a breakneck pace for decades, with the expansion accelerating in the past 15 years.
The resulting system is inherently highly unstable and becoming ever more so. This is why we are seeing so many "squirrely" market moves, to paraphrase Bill Fleckenstein. It also means that there is always a certain latent crash danger in the air. While crashes normally don't happen while money supply growth remains elevated, there are other factors that may temporarily trump this fact. Consider for instance that if the idea that forced selling has taken place is correct, there might conceivably come a day when one or several large leveraged players find themselves in severe trouble. In that case the recent 2008 experience may trigger the herding effect much faster than would normally be the case, due to the lesson learned at the time, which was: "it is best to panic early, and do so in size".
Charts by: BarCharts, StockCharts, Saint Louis Federal Reserve ´Research