Should you buy into the bond rally? Yes, says James Ferguson, there's plenty of upside left. But Dr Peter Warburton argues we're in a bubble - and it's time to sell.
Is the bond market rally a bubble? No. A bubble occurs when an investment story, which starts off with sound economic fundamentals behind it, develops its own momentum and, after a while, its own set of justifications for existing usually along the 'this time it's different' line.
The cyclical bond bull market has been too short and the structural bull too long for this to qualify as a momentum bubble. The most recent bull phase only began four months ago, at the start of April, having spent the prior 15 months selling off. The long-run bull market, on the other hand, has been going for 29 years. Government bond yields have been on a steady downtrend since as far back as September 1981. Another sign of a bubble is when an asset rises on 'irrational exuberance' (for bonds, which are seen as a safe haven, this would translate into 'irrational dismay').
Subscribe to MoneyWeek
Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE
Sign up to Money Morning
Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter
Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter
So, is the bond market's dismay irrational? To look at stockmarkets, at least until the peak in April, you'd have to say 'yes'. However, if you look back through history, in most cases where the stockmarket and the bond market adopted differing world views, the bond market was proved right. Stockmarkets get too caught up in individual stocks and recent profit history. But you ignore what the bond market is telling you at your peril. The bond market told you to buy stocks three months before the rally began in March 2009. It told you to get out again in April this year, when the ten-year benchmark US Treasury yield failed to break the 4% ceiling. Since then its warnings have become increasingly strident.
Equity bull markets famously 'climb a wall of worry'. In the case of bond bull markets, it's more a slow crushing of optimism. As each non-believer comes round to the idea, they capitulate and buy in. The bull market climbs higher and yields fall further, until finally everyone believes and there are no more new buyers left. This final stage is often when the bubble is formed, as prices keep rising even after the original fundamental cause has faded away. So the fact that there are still so many bond nay-sayers around makes it less likely that this is a bubble, not more so. It's only once everyone's a bond bull that we should be worried.
And it's always tempting to call any market that doesn't go the way you expect a bubble. It implies there's no fundamental reason for it happening, which lets you off the hook for not seeing it coming. We've all done it. However, I did see this one coming, as I most recently reiterated in a MoneyWeek debate with another bond bear back in February. And since then the surge in bond prices has been electric.
The total return on the US 10-year+ Treasury has been around 20%, against a loss of 6% year-to-date in the S&P 500. The return on the 30-year Treasury has been more than 30%. Here in Britain, the ten-year gilt total return has approached 14%, while the 30-year has given you almost 20%. Meanwhile, the FTSE 100 has also lost 6% so far this year.
These are such strong gains that I'm sure we're overdue a correction. But that does not mean we're in a bubble, nor that the bond bull market is over.
Another concern of the 'bond bubble' brigade is that bond yields are too low to tempt any rational investor. Well, I hate to point out the obvious, but there is no positive number (for which read 'yield') that can't halve. So there's always big upside potential, even at very low yields. In fact, the lower the yield, the bigger the gearing, especially for long-dated bonds. Here's why.
The ultimate in long dated bonds is a perpetual, with no maturity date. The price of such a bond doubles if the yield halves from 10% to 5%. It would then double again if the yield halved to 2.5% from 5% and again if the yield went down to 1.25%, and so on. So while it took a 5% point move in the yield to double the price first time around, it only took a 1.25% point move the final time. So even at the current 2.84%, ten-year gilt yields still offer great upside potential. You just have to look at ten-year German Bunds, which yield 2.15%, to see that.
Special FREE report from MoneyWeek magazine: Don't be fooled - house prices will fall again!
- Why UK property prices are set to collapse by 30%
- When it will be time to get back in and buy up dirt cheap property
So, if the bond bull market hasn't been going long enough to be a bubble and yields could theoretically still offer substantial upside, can it see off the other bubble accusation: that it's not fundamentally justified? Well, I've already argued the fundamental case back in February, before the big rally in bonds. So obviously I believe the case is strong. It all boils down to whether or not this is a normal cycle recovery. The academic literature is littered with studies on post-bank crisis resolution. They all come to the same conclusion. Until banks are finished shrinking their loan books, private-sector growth is always anaemic at best. The only hopes for a positive GDP trend are for export-led growth, inventory adjustment or deficit-funded public sector 'growth'. And only the first of these is sustainable.
We are just coming to the end of the inventory adjustment phase of the cycle. This boosted US GDP growth (quarter-on-quarter, annualised) to 5% in the fourth quarter of last year. But since the end of April, for example, the closely watched US durable goods orders data has been negative. There are increasing fears about just what sort of underlying real growth lies behind the boost from the end of inventory de-stocking, not least because one investment bank has concluded that the end of the Bush tax cuts will knock 1.3% off headline GDP growth next year. One clue comes from real final sales, which reflect real GDP less the distortions caused by changes in inventories. These are growing at just 1.2%. Record fiscal and monetary stimulation appear to have achieved almost nothing.
That's because in a fractional banking system like ours, growth in broad money supply is driven by bank loan growth. All other things being equal, when bank lending shrinks, so does the money supply, and thus the economy too. To prevent this, the Bank of England initiated £200bn of quantitative easing (QE) worth about 14% of GDP. Yet broad money supply ('M4 excluding intermediate OFCs') was only 1.2% higher in the second quarter than in 2009. That's the lowest rate of growth since the 1930s. It suggests that without QE, money supply might have shrunk by at least 6%-7%. Ireland, for example, having no independent central bank to try QE, has seen its broad money (M3) shrink by 14% and its nominal GDP by more than 20%. QE is not a reason to fear inflation. It's the reason we aren't in a full-throttle deflationary depression and the bond market knows it.
So the bond market hasn't risen far enough, or for long enough, to be a momentum-based bubble. Nor is its potential upside constrained, nor is it based on an irrational macroeconomic premise. So let's look at how much higher the fundamentals could yet propel it. The real yield what the long bond yield pays above the rate of inflation has averaged about 2%-3% over the years. So today's 2.84% yield on the ten-year gilt is bang in line with expectations for very low inflation, but not yet deflation.
Forget the headline UK Retail Price Index (RPI) inflation figure of 4.8%. This is distorted on a year-on-year basis by low comparables. If you annualise the two-year RPI, the inflation rate falls to just 1.8%. Barring the odd month in the aftermath of the dotcom crash, that's the lowest since 1961. Core inflation in the US and eurozone is already nudging 50-year lows.
Since 1998, 30-year gilt yields have been locked in a surprisingly tight band of 4%-5%. During this 12-year period, while 30-year gilts yielded 4.5% on average, RPI inflation averaged 2.6% and Consumer Price Index inflation 1.9%. If underlying RPI is really 1.8%, that historical relationship suggests a yield of 3.7%, lower than today's 30-year gilt yield of around 3.8%. Rather than being in a bubble, the bond actually looks fundamentally undervalued. That shouldn't come as a surprise. If multi-decade low levels of inflation, a credit system that looks likely to be impaired for another three to four years, the lowest money supply growth since the 1930s, and an economy that looks dangerously close to double-dipping aren't justification for a record-low bond yield, I hardly know what could be.
Japanese ten-year government bonds (JGBs) yielded 5% in late 1994 when bank loan growth first went to zero. That's very similar to the 4.5% that ten-year gilts were yielding at the end of October 2008 when bank lending here peaked out. But by June 2003, after lending had been shrinking for six to seven years straight and even though Japanese government debt had hit almost 200% of GDP (UK debt is currently less than 70%, though rising fast) the JGB yield was less than 0.5%; a tenth what it had been eight-and-a-half years earlier. Countless investors lost their shirts along the way betting that was a bubble too.
James Ferguson is Head of Strategy at Arbuthnot Securities and writes the Model Investor newsletter.
As a chart of the US ten-year Treasury yield (above) clearly shows, bonds are not in a bubble. A bubble is when yields spike down below the trend dramatically (as in Japan in 1998 see below). But bear in mind that Japan's bubble was caused by almost exactly the same set of circumstances that we now face in the West.
The bond market feeding frenzy
Dr Peter Warburton
Since mid-April, the yield on a US ten-year Treasury has fallen by 1.27 percentage points, equivalent to a one-third fall. And because elite government bond markets are very closely correlated, US bond market behaviour has been copied around the world. In Britain, for example, ten-year gilt yields have had an 87% weekly correlation with US bond yields since the credit crisis erupted in August 2007. That correlation has risen to 95% since the start of this year.
I'm going to look at why bond yields have shrunk and why this tells us next to nothing about the global inflation-deflation debate. Unlike James Ferguson (above), I believe that this sharp fall in government bond yields has shaky foundations and is very vulnerable to a reversal. Perhaps, rather than calling this a bubble in government bond prices (equivalent to an inverted bubble in their yields), it is more accurate to say that there has been a feeding frenzy in government bonds. A bubble bursts when its infrastructure fails. A feeding frenzy on the other hand, ends when the food supply is exhausted. In this case, the food consists of very short-run bond trading opportunities.
The four main ingredients to the story
1. The false market in US mortgage-backed securities
The Bank of England's asset purchase scheme (its quantitative easing programme) focused mainly on gilts. The US Federal Reserve, on the other hand, focused on mortgage-backed securities (MBS) instead. These are bonds issued by government-backed groups Fannie Mae and Freddie Mac. The bonds are issued against bundles of mortgage loans that conform to certain criteria regarding the maximum loan size, type of loan and the creditworthiness of the borrower when the loan was made. The Fed's purpose in buying these MBS was to keep their yields down and so spread a culture of cheaper mortgages throughout America. In theory, the Fed's asset purchases would reduce monthly mortgage payments on existing loans and enable borrowers on fixed interest rates to refinance at lower mortgage rates. US consumers would then have more disposable income and so would spend more on goods and services. The strategy has been rather less successful than hoped. It has definitely eased the burden of mortgage payments for millions of borrowers. But it has failed to bring relief to the borrowers in greatest need because their delinquency stopped them from refinancing.
However, the real significance of the Fed's $1,250bn of MBS purchases lies in the market distortions it has left behind.
In trying to complete its asset purchase programme by the end of March, the Fed sucked up too many of the longer-duration MBS bonds. So fixed-income investors who would once have been able to replace this duration in the Asset-Backed Securities market were driven towards the US Treasury market instead.
In other words, the Fed's intervention in the MBS market brought a new group of buyers into the Treasuries market.
2. The wonderful world of bond-market convexity
Forgive me for getting a little technical here. Convexity is the rate at which bond market duration the sensitivity of bond prices to interest rates changes as a function of yields. Convexity has value. The greater the convexity, the more the bond price gains as interest rates fall and the less the bond price loses as interest rates rise. The convexity of the US Treasury yield curve between five years and ten is close to the record-high seen in 2008. To cut a long story short, bond investors can profit by betting that the yield curve between these two points will flatten out. Some would argue that, as long as there are convexity gains to be made in duration-neutral trades, long-term bond yields can keep falling. But this has nothing to do with the prospects for the economy or the price of a pint of beer: it is an internal dynamic of a sophisticated bond market. As ten-year yields fall relative to five-year ones, the demand for ten-year bonds grows. But this dynamic can operate with equal force
in reverse.
3. A turnaround in investor risk appetite
After the broad US equity market had managed a year of impressive gains from its March 2009 low, there was an increasing likelihood of a technical setback. The fracture of the eurozone sovereign debt markets in March 2010 were one major factor in convincing investors to pull out of the equity and commodity markets. The end of QE in Britain and America was also significant. The transition from "risk-on" to "risk-off", as investors looked for 'safe-haven' investments, helped boost benchmark bonds.
4. Exaggerated fears of global economic decline
The superficial similarities between the Greek sovereign debt trauma and the Lehman Brothers collapse of September 2008 inspired some pundits to predict a crisis of the eurozone economies and, by extension, the global economy this year.
No such crisis has occurred. Nor should we fear one, as ultra-easy credit policies continue to propel an acceleration in nominal GDP around the world. Instead, erratic quarterly patterns of GDP growth have fuelled scaremongering over the near-term economic outlook in various countries at different times. After the eurozone reported robust growth in the second quarter, the baying hounds turned towards the softer-than-expected growth in the US economy. However, net exports have taken a huge bite from this second quarter figure and this is very unlikely to be repeated in the third quarter. The US is suffering particularly from an ongoing slump in its housing market and from its weak state and local finances. But even these setbacks do not add up to a double-dip scenario. Unfortunately, the US Fed has indulged these fears with its announcement of a resumption of Treasury debt purchases from the stream of maturing MBS debt. Economic fears are being whipped up as the justification for bond-price movements that have no fundamental basis.
Does cheap credit work?
Advocates of government bonds, especially at these shrunken yields, have persuaded themselves that cheap credit is unable to revive the global economy. But this goes against both economic logic and human experience. Just because historically low short-term interest rates have failed to revive private-sector credit demand in the so-called advanced economies over the past 18 months, doesn't mean that we let this experiment run for another 18 months, let alone three to five years. Cheap credit isn't reaching the willing hands that would take it because banks and other intermediaries are demanding a huge profit margin for their services. In some cases, banks are doing this because they don't wish to lend at any price; in others, they're eager to rebuild their capital and sense that the forces of competition are weak. But in the past three months there are signs that cheap credit is reaching distressed corporate borrowers. New channels of lending are opening up that will bypass traditional lenders or compete with them for personal customers. Mortgages are becoming more affordable and available as lower bond yields feed through to cheaper fixed-rate loans.
Not even a broken banking system will prevent cheap credit getting through to our economic system, given time. As private-sector borrowing recovers, banks will relax their lending attitudes and substitute mortgages and consumer loans for purchases of government debt. The annual growth rate of money-value purchases in the US and Europe will rise from 3% to 5% to 7% and so on. The prospective total returns to global equities will improve accordingly, while the attractions of government bonds will erode. Investors will rediscover their appetite for risk, and the consensus that's enveloped US Treasuries, gilts, bunds and JGBs will be disturbed as the convexity trade unravels and hedge funds look somewhere else for kicks. US and UK benchmark ten-year bonds yields could rise 0.5% in a month. Perhaps even in a week.
Peter Warburton is director of Economic Perspectives Ltd.
This article was originally published in MoneyWeek magazine issue number 502 on 3 September 2010, and was available exclusively to magazine subscribers. To read more articles like this, ensure you don't miss a thing, and get instant access to all our premium content, subscribe to MoneyWeek magazine now and get your first three issues free.
