At extreme junctures the crowd is most prone to extrapolate the unrepresentative present into the indefinite future. The result is that the more intense is today’s craze, the more prominent are the intellects that insist it’ll last. From this point of view, it’s hardly surprising that the longer is the period of time during which rates of interest fall and the lower the level to which they descend, the greater is the number of people (particularly of influential and vocal people) who believe – and the more fervent is their conviction – that rates will certainly remain extremely low and probably fall even further.
I suspect that the conviction of today’s mainstream that “rates will stay very low for very long” doesn’t just reflect their worship of abstract models and fresh-and-blood central bankers, and their disregard of the historical record: it belies their hubris that the future is foreseeable and that our rulers, at least, can foresee it. I fear it’s the old delusion (“it’s different this time”) in a new guise. When the crowd celebrates the alleged exceptionalism of the times, it pays the genuine investor to investigate contrary possibilities. Rather than parrot the mantra that “rates will undoubtedly remain low,” it’s more fruitful to consider the contrarian question: “what might happen when they rise?”
Is it really different this time? The historical record demonstrates convincingly that in one respect it undeniably is. Equally, there’s no reason to think that it’ll remain so – and several to conclude that it won’t. How to ascertain whether the present differs significantly from the past? The most sensible way, it seems to me, is to analyse long series of valid and reliable data. What’s the best way to consider central banks’ discount rates and the yield of “risk-free” sovereign debt? I believe it’s to put them into the longest possible context. What institution provides this context? The Bank of England does.
Using data from the BOE’s web site, Figure 1 plots the BOE’s base rate since October 1694 and the yield of HMG’s long-term bonds since June 1703. It’s analogous to the Fed’s discount rate; as such, it strongly influences longer-term rates including mortgages, corporate bond rates etc.
Bank of England’s Base Rate and HMG’s Long-Term Bond Yields, October 1694-December 2014
Two things in Figure 1 are clear: never has the Bank of England pushed its base rate (1) remotely as low and (2) nearly as quickly as it began to do late in 2008. In October of that year, the base rate was 5%; in November, the Bank cut it to 3%; in December, to 2%; in February 2009, to 1%; and in March 2009, to 0.5%. Within six months the Bank did something that it had never done before: it slashed its base rate by 90% (that is, (0.005 – 0.05) ÷ 0.05). Until February 2009, it had never set it below 2%; since March 2009, it’s pinned it at a mere one-quarter of that level. The period since early 2009 has no historical precedent; in that sense, it is indeed fundamentally different this time.
Although the present is unprecedented, I suspect that the future (which, alas, I can’t foresee) won’t be: that is, since June 1822 the base rate has fluctuated erratically around its long-term mean – which from 1694 to October 2008 was 4.82%, and for the entire series (1694-2014) is 4.77%. What’s a “normal” base rate? Figure 2 provides a range of means from which to choose. Whichever you prefer, the “normal” rate greatly exceeds the average since 2009.
Bank of England, Average Base Rate during Various Intervals, 1694-2014
Sometimes the base rate soared far above its overall mean: in particular, during most of the 1970s and 1980s it exceeded 12%. At other times it sagged well below its mean: during the 75 years after 1835 it averaged 3.7%, and during the 1930s and 1940s it averaged 2.3%. Notice, however, that during these below-average periods both base rate and bond yield frequently spiked well above their overall averages.
Slightly above-average and slightly below-average base rates have prevailed for great lengths of time. Most notably, the rate remained constant at 4.5% for 17 years (i.e., from June 1699 to June 1716) and at 5.0% for an astounding 103 years (i.e., from April 1719 to May 1822). The bulls can indeed point to extremely long periods during which the Bank of England never changed its bank rate. They can also cite extended intervals of below-average rates.
Two inconvenient (for the bulls) facts remain. The first is important enough to repeat: at no time has the base rate ever fallen as far and as fast as it did in October 2008-March 2009. Second, at no time has it remained as low as it has since March 2009. A slightly below-average rate existed – albeit hardly continuously – for ca. three-quarters of a century after 1835; but the record low which has prevailed since March 2009, or indeed anything like it, has never done so.
Hence our impudent question to today’s bulls: if since 1694 it’s never been different (in the sense that you intend), then why – apart from the lame reason that central banks must suppress their base rate in order to support the bacchanalia – should anybody expect that the next 70 (or even 20 or 10) years will be different from any other corresponding period between 1694 and 2008?
Eschewing the bulls’ exuberance and bearing in mind that since 1822 the Bank of England’s base rate hasn’t over time conformed to a rigid, replicable and hence reliably and profitably predictable pattern, we can express our scepticism more specifically. Figure 1 shows that the base rate has tended, erratically but eventually, to regress from extremes towards its overall mean. So too the long-term bond’s yield: lower yields eventually follow above-average ones, and higher yields follow below-average ones. If things aren’t, after all, different this time, then it’s reasonable to expect that at some point – alas, neither I nor you or anybody else knows when, but it’ll likely take us all by surprise – not only will rates rise: they’ll also revisit their super long-term average. When they do, I suspect that there will be hell to pay.
The Decrease of Rates Since the Late-1970s/Early-1980s
Why, generally speaking in most Western countries, have rates of interest fallen almost continuously – and cumulatively massively – since the 1980s?
Figure 3 plots the yield of 10-year U.S. Government and Australian Government bonds since July 1969 (when directly-comparable data became available). Why 10-year securities? They’re a mid-point between the longest-dated “risk-free” debt (the 30-year U.S. Treasury bond) and its shortest-term counterpart (Treasury bills whose duration is one year or less). And as we’ll shortly see, the 10-year has become the primary means whereby the Treasury finances the U.S. Government’s deficit.
The Yield of Ten-Year Government Securities, Australia and U.S., 1969-2014
Both series in Figure 3 tell much the same story: yields rose in the 1970s, peaked in the early 1980s and thereafter have fallen almost continuously. In the U.S. since 1969, the ten-year yield has averaged 6.8% and in Australia it’s averaged 8.5%. Treasuries’ average yield since 1969 has differed little from British gilts’ (which isn’t shown); the average rate of ten-year Commonwealths, on the other hand, exceeds the gilts’ and Treasury 10-years’ average.
The most significant (statistically, it explains more than one-third of the total variation since 1969) proximate cause of changes of rates of interest is the deceleration of the Consumer Price Index (CPI). Figure 4 plots the annualised increase of CPI in Australia and the U.S. since 1969. Like the 10-year yield, so too the CPI: it rose rapidly in the 1970s, peaked in the mid-1970s (Australia) and early-1980s (U.S.) and thereafter has decelerated almost continuously. In the 1970s, CPI averaged 6.9% in the U.S. and 9.2% in Australia; in the 1980s, it averaged 5.6% in the U.S. and 8.4% in Australia; in the 1990s, it averaged 3.0% in the U.S. and 2.5% in Australia; and since 2000 it has averaged 2.4% in the U.S. and 3.0% in Australia.
CPI (Quarterly Observations), Australia and the U.S., 1969-2014
Other things equal, an acceleration of CPI causes yields to rise and a deceleration causes them to fall. Bonds’ yields rise in response to an increase of CPI (more precisely, yields rise when lenders expect that the cumulative CPI during the duration of their loan will accelerate) because the increase of CPI erodes the future purchasing power of bonds’ principal and payments of interest. In the 1970s, as CPI accelerated, long-term investors’ “real” (net of CPI) returns from 10-year Treasuries fell. These and other investors began to expect that CPI would remain high and rise higher. In order to protect themselves against this erosion and to generate a positive rate of return, lenders demanded much higher rates of interest; accordingly, Treasuries’ yields steadily rose.
In the early 1980s, investors feared that the rise of CPI might return to the levels (above 10% per annum) it had scaled just a few years before; accordingly, and in order to protect themselves against this possible severe erosion of their purchasing power and to generate a positive rate of return, they demanded compensating very high rates of interest. Conversely, once CPI’s rise decelerated towards 3% per annum and investors became convinced that it’d stay there, they became willing to lend at far lower rates.
With that thought in mind, let’s consider matters as they stand today. In the year to 30 September 2014, the yield of the 10-year U.S. Treasury has averaged 2.53% and CPI has averaged 1.75%. Do you see the risk that its purchasers face? Let’s assume (to the crowd it’s an article of faith) that CPI accurately measures prices and their changes over time. Also assume (I doubt it, but who knows?) that this rate of change of CPI will remain constant during the next decade.
Given these assumptions, consider as an example the investor who purchased a 10-year Treasury (again, for simplicity assume a purchase price of $100) yielding 2.53% in 2014. After 10 years, if CPI increases at a constant rate of 1.75%, in nominal terms the investor will receive total payments of interest of $25.30; net of the effect of the very low CPI, he’ll receive $22.97; and when Treasury redeems the bond in 2024, he’ll receive principal whose nominal value will be $100 but whose purchasing power is $83.80. After ten years under these circumstances, the investment’s purchasing power increases from $100 in 2014 to $106.77 (i.e., $83.80 + $22.97) in 2024. That’s a “real” compound rate of return of 0.66% per year.
Even given rosy (I’d call them “aggressively optimistic”) assumptions, the real rate of return the investor can realistically expect is less than 1% per year. Is that sufficient compensation for the risks he takes? What if even slightly less optimistic conditions prevail? If, for example, during the next decade CPI averages 2.4% (its average since 2000) the “real” rate of return decreases to -0.1% per year. Any average rate of CPI greater than 2.4% per year generates increasingly bigger losses. Under current circumstances, Treasuries don’t seem to offer “risk-free” return; if anything (the phrase is James Grant’s), they promise return-free risk! Who’d buy these securities under today’s unappealing circumstances? In order to protect against the risk that an unexpectedly high rate of increase of CPI erodes the purchasing power of their capital, and to generate an acceptable “real” rate of return, I suspect that at some point sooner than 10, 20 or 70 years lenders will demand more adequate compensation. What if, for example, they demand a rate of return (net of CPI) of 2% per year? Assuming that during the next decade CPI rises 2.5% per year, purchasers of 10-year Treasuries must demand a yield of 4.6% – which is much closer to its long-term historical average, and almost double its average during 2014.
Why Has CPI Decelerated and Remained Stable?
Since the early 1980s, rates of interest have fallen primarily because the CPI’s rate of increase has decelerated. Why has its rate of increase slowed? Two sets of reasons: the first is largely market-based and therefore beneficial, legitimate and maintainable; the second is wholly non-market and hence harmful, illegitimate – and, I suspect, untenable.
-Three Largely Market-Based Reasons
Deng Xiaoping – China’s leader from 1978 until his retirement in 1992 – symbolises the first reason. Of course, he didn’t single-handedly cause the CPI’s rate of increase in Australia, the U.S. and elsewhere to slow. Beginning in 1979, however, he commenced a series of economic reforms which have expanded and accelerated over the years and whose cumulative scope and consequences have no precedent. Hundreds of millions of Chinese peasants began to manage and own the land they cultivated and to sell their output on domestic markets. At the same time, China’s economy quickly opened to foreign trade and investment. As a result, since the late-1970s its imports of food have plummeted and its exports of ever more advanced goods (produced by scores of millions of people who’ve migrated from the countryside to rapidly-growing cities) have exploded. Never before, surely, have so many risen so quickly from such abject poverty.
China and other “emerging economies” have become the world’s biggest producers and primary suppliers to Western nations of a vast number of ever more advanced goods which Westerners were once the leading producers. Not just because their labour has been quite inexpensive (but progressively less cheap over time) by Western standards, but also because their technology has usually approached and sometimes exceeded Western standards, the emergence of China and other developing countries as major forces in the international economy has increased the supply of goods and services relative to the supply of money. Developing nations’ rise to prominence in a more integrated (“globalised”) world economy has (and other things equal) caused CPI in Western countries to increase less rapidly than it otherwise would.
Bill Gates symbolises a second reason. Neither he nor Steve Jobs or any other CEO single-handedly caused its rate of increase to slow. They do, however, epitomise the IT revolution, dating roughly from the late 1970s, which has enhanced the productivity of production, transport, marketing and distribution. Better IT enables a given amount of output to occur at lower cost (or more output at the same cost). Other things equal, these advances of productivity increase the supply of goods and services relative to the supply of money, and thereby place downward pressure upon the prices of goods and services. Nowhere has this phenomenon been more apparent than in the computing and IT industries.
Sir Roger Douglas and Paul Keating (the Treasurer, not the PM) provide a third reason. As much as anybody, and more than most, each promoted reforms that commenced in all Anglo-American countries (but later and much less thoroughly in Western Europe) in the late 1970s. These reforms, which by the late 1980s had become near-orthodoxy, have enhanced domestic competition for goods, services and labour. More vigorous competition, in turn, has (again, other things equal) tended to place downward pressure upon prices.
-Two Non-Market Reasons: Heavy Purchases by Foreigners and the Fed
How does the government affect rates of interest? Specifically, how does its fiscal policy (namely the size of its budget’s deficit or surplus) affect rates? “The conventional view,” say Douglas Elmendorf and Gregory Mankiw (Government Debt, Federal Reserve Board, Finance and Economics Discussion Series, 1998-1999), “is held by most economists and almost all policymakers. According to this view, the issuance of government debt … ‘crowds out’ capital and reduces national income in the long run.” Fifteen years later, that remains the mainstream’s assessment. It’s not wrong: in plain English, and assuming that the deficit is financed solely from domestic savings and holding constant the policy of central banks (which are vital caveats which we’ll revisit), the government finances a deficit of (say) $1m by borrowing $1m – that is, by issuing to the general public securities of whose total face value is $1m.
Private investors exchange this amount of cash, which the government then spends, for the government’s newly-issued bills, bonds and notes. The cash which private investors use to buy these debt securities cannot, of course, simultaneously be put to other uses – such as the purchase of private sector bonds, shares and other securities. The government’s deficit thus reduces the total quantity of investable funds but leaves unaffected private entities’ demand for these funds. The greater is the size of the government’s deficit relative to the private sector borrower’s demand for funds, the more the government’s demand for finance “crowds out” the private sector’s.
Like a game of musical chairs in which the number of contestants exceeds the number of chairs, the unchanged private demand for a diminished (after the government’s issue of securities) supply of funds increases competition for these funds. How do private borrowers ensure that they’ll win a seat (i.e., obtain funds) before the music stops? They offer more attractive terms and conditions – namely higher rates of interest – to lenders. Unless domestic savers suddenly increase their propensity to save, or the private sector’s demand for investable funds subsequently decreases, or the government finds some alternate source of finance, given the aforementioned assumptions a government which consistently runs deficits will place upward pressure upon rates.
The phrases “America’s national debt” and “total face value of the U.S. Treasury’s outstanding securities” are synonyms. Figure 5 stratifies the U.S. Government’s debt obligations according to their duration: the Treasury’s bills are short-term debt (repayable less than one year); its notes are medium-term (repayable in 1-10 years); and its bonds are long-term (payable in 11-30 years).
The U.S. Government’s Debt Obligations, Total and Major Components ($US Trillion), 1969-2014
Debt has grown from $259 billion in 1969 to $17.829 trillion in mid-2014. That’s a compound rate of growth of 9.9% per year. The size of America’s economy, measured by Gross Domestic Product (GDP) in nominal dollars, has increased from $1.02 trillion in 1969 to $16.77 trillion in mid-2014. That’s a compound rate of growth of 6.7% per year. If only parenthetically and rhetorically, it’s nevertheless worth asking: whether for an individual, household, corporation or government, can debt indefinitely grow more quickly than gross income (revenue)?
At first glance, the analysis to this point disconfirms the contention that large and rising deficits place upward pressure upon interest rates. Not only has the U.S. Government generated deficits virtually without exception since 1969: its cumulative deficit – that is, the national debt – has grown exponentially. Yet these large and rising deficits clearly haven’t impeded the almost continuous decrease of rates of interest since the early 1980s. Much lower rates have accompanied far bigger deficits. How can this be? Recall the two assumptions: first, domestic savings provide the sole source of deficit finance; and second, central banks’ policies remain constant. In 1969, these assumptions corresponded roughly to reality; by 2014, however, both had long ceased to do so.
Figure 6 disaggregates America’s national debt into four categories of ownership (directly-comparable data are available only for the years since 2000). In 2000, private American entities (such as insurance companies, investment funds, etc.) owned ca. one-quarter of all Treasury securities. This percentage had fallen from more than one-half in 1969. By 2014, this percentage fell further, to ca. one-fifth. Private American investors, in other words, were once the U.S. Government’s primary source of finance; but since at least 2000, they no longer do so. The government’s deficits no longer “crowd out” private investors because it now has other – and more significant – sources of finance. Who are they? Foreigners – specifically, foreign central banks.
Who Owns the U.S. Government’s National Debt? Major Components ($US Trillion), 2000-2014
In 1969, foreigners owned a negligible percentage (less than 5%) and amount ($6 billion) of Treasuries; of these, private (including corporate) investors in Britain, Canada, France and Germany comprised most of this amount. By 2000, foreign holdings had skyrocketed to ca. $1 trillion (15% of the total) and in 2014 to $6 trillion (33%). Particularly noteworthy are China and Japan: in each country in mid-2014, central banks owned ca. $1.4 trillion of Treasuries (relative to their populations, South Korea and Taiwan are also big holders).
Why have foreigners, particularly the People’s Bank of China and the Bank of Japan, become (apart from the Fed and U.S. Government) the biggest holders of Treasuries? Asian central banks, using domestic currency they’ve conjured out of thin air, have long purchased huge amounts of $US. These transactions have decreased the demand for their currencies relative to supply; they’ve also boosted demand for $US. As a result, central banks have depressed these currencies’ rates of exchange vis-à-vis the $US – and cheapened the price in $US of their exports to the U.S. These central banks have then used significant quantities of their vast $US reserves to buy Treasuries. These purchases, in turn, have accommodated America’s skyrocketing budget deficits. Because they’ve become the biggest buyers of Treasuries, Asian countries’ strong demand has boosted Treasuries’ prices and thereby suppressed their yields. Asian central banks have become and remain such eager purchasers of Treasuries that the U.S. Treasury has been able to sell them under increasingly advantageous (to the Treasury) terms and conditions. Yet these terms and conditions also suit Asian central banks: since Treasuries’ yields provide the benchmark for other rates, Asian central banks’ voracious appetite for Treasuries has also suppressed the yields of sovereign and corporate debt in China, Japan, Korea, Taiwan and elsewhere.
Since 2007, foreigners and the Fed have comprised a strong majority – ca. 70% – of the total; and of this total, moreover, foreigners have demanded the lion’s share. Since 2007, in other words, the U.S. Government has depended primarily upon the Fed and major foreign central banks to finance its growing deficits. These central banks’ strong demand for Treasuries has increased their prices and thereby suppressed their yields.
What Have the Fed and Other Central Banks Wrought?
What happens if and when these central banks decide to decelerate the rate at which they accumulate Treasuries – or even decrease their holdings? Assuming that it continues to run big deficits, either the Fed must buy (monetise) more debt or the U.S. Government must offer significantly better terms and conditions – that is, materially higher rates of interest – in order to entice other borrowers. What’s the likelihood that China’s demand for U.S. Treasuries will abate? For a decade, studies have found “evidence that some types of investment are becoming excessive in China, particularly in inland provinces.” In these regions, economic activity has become more dependent upon superabundant and hence wasteful capital investment, whose object is the export of consumer goods and whose impact is short-lived,
In short, if China’s economy successfully evolves from heavy
concentration upon capital investment and the export of goods towards
greater focus upon domestic consumption of goods and services, then its
dependence upon exports, the People’s Bank of China’s need to suppress
the currency’s rate of exchange – and thus the Bank’s need to buy U.S.
Treasuries – will all lessen. Perhaps something along these lines has
begun to occur: in November 2013, the PBC announced that it intends to
decelerate its accumulation of foreign currencies (including $US).
According to Seeking Alpha (People’s
Bank of China Announces End of U.S. Treasury Buying,
22 November 2013), its purchases of Treasuries will therefore slow.
Janet Yellen points to today’s tiny rates of interest as evidence that
the Fed’s “stimulus” since 2008 has succeeded. So does Ben Bernanke (Bernanke
Tells U.K.’s King: “We Saved Our Economies”,
MarketWatch, 29 December 2014). In sharp contrast, I regard them
as dangerous experiments whose consequences (namely reckless lending,
borrowing, investment and consumption) will cause havoc when rates
revert towards their long-term means. Easy money today merely defers
harder decisions to tomorrow. Am I crazy or is Janet? Am I too cautious
or is the mainstream too confident? Are the bulls right to boast that
rates won’t rise? Or, deep down, do they really fear that they
mustn’t rise because the consequences will be so dire?
Little – least of all our rulers’ hubris – surprises Bill Bonner, president of Agora Publishing and co-author of Empire of Debt: The Rise of an Epic Financial Crisis (John Wiley & Sons, 2006). “The Fed’s EZ money policies,” he says,
Interventionists’ attempts to plan the economy centrally have repeatedly
failed dismally. Have we any reason to expect that “it’s different this
time” – that is, that ZIRP, QE and all the rest won’t, like all other
interventions, trigger unintended negative consequences (first economic
stagnation, finally financial crisis)?