- The lessons from the 2013 taper tantrum are relevant for bond investors today, but as circumstances are different we do not expect a similar outcome.
- Central banks have been very clear that adjustments to balance sheet policies do not signal a shift in the broader accommodative stance, and so far have been effective at communicating that message.
- The negative correlation across bonds and risk assets is likely to hold, much like it did over most of 2013.
- The starting point matters for asset prices, and EMs are notably at a better starting point than in 2013.
- Over the longer term, we believe markets will respond to inflation and growth, neither of which is likely to be impacted very much by upcoming adjustments to central bank balance sheets.
When a signal from the Federal Reserve (Fed) regarding its plans to “step down” asset purchases in 2013 led to a sharp sell-off in markets—as bonds, risk assets, and especially emerging markets (EMs) all moved down in price simultaneously—the taper tantrum of 2013 had arrived.
Now in 2017, central bank balance sheets are again top-of-mind for bond investors. The Fed expects to adjust its reinvestment program later this year, with an eye toward eventually shrinking its balance sheet by $1-2 trillion. As discussed in a recent paper by my colleague Bonnie Wongtrakool on Fed Balance Sheet Normalization, this would be yet another step in the Fed’s slow and gradual exit from extraordinary monetary policy. The European Central Bank (ECB) is a few years behind in the process but nonetheless has signaled a desire to move in the same direction.
Large central bank balance sheets have been a prominent feature of the post-crisis landscape, and it’s understandable that bond investors are watching these developments warily. A few have even raised the concern that apparently small changes in balance sheet policy can have an outsized impact on markets, as was the case in 2013.
Given this recent experience, some wariness about Fed moves—and the market’s potential reaction—is understandable. But, is the taper tantrum really all that relevant for understanding the current landscape?
Before answering that question, we think it’s useful to keep in mind the following four lessons from 2013:
- Markets mispriced the Fed’s intentions in 2013, especially in regard to short-term interest rates.
- The negative correlation between bonds and risk assets was upheld over the course of 2013, although it did break down for one month at the start of the taper tantrum.
- The starting point matters in terms of the subsequent market reaction. For certain asset classes, today’s starting point is substantially different than in 2013 (notably, for EMs the starting point is very different).
- Growth and inflation have more influence on yields than central bank policies in the current environment.
Considering these four lessons, we suggest that the taper tantrum of 2013 is not the best playbook for investors concerned about changes to central bank balance sheets.
In contrast to 2013, central banks have been and will continue to be very clear that their policies of broad accommodation remain in place, even as they tinker around the edges. Just like in 2013, the negative correlation between bonds and risk assets is likely to hold, with asset prices responding more to fundamental value rather than central bank actions. And although central banks continue to forecast firmer inflation, adjusting their balance sheet policies in anticipation is unlikely to make it so.
This note expands on the four lessons listed above, and by doing so seeks to illustrate the fallacy of drawing parallels between 2013 and the current environment.
Markets Mispriced the Fed’s Intentions in 2013
In May 2013, then-Chairman Ben Bernanke said that the Fed could “step down” the pace of asset purchases “in the next few meetings.” There was no hint or suggestion in his comments that the Fed was also considering changes to the interest rate policy. In fact, at the beginning of the same testimony, Bernanke emphasized that the Fed’s forward guidance in regard to interest rates “underscore[d] the Committee’s intention to maintain highly accommodative monetary policy as long as needed.”
Yet, market participants struggled to make the distinction between changes in asset purchases and the broader accommodative stance. In the immediate aftermath of Bernanke’s comments, the market repriced its expectation for the future path of short-term interest rates from one hike to four full hikes by mid-2015 (Exhibit 1). And this repricing was in spite of the fact that Fed officials were pushing back very hard and reasserting the forward guidance at every opportunity. For example, in July 2013, New York Federal Reserve President Bill Dudley said that “a rise in short-term rates is very likely to be a long way off.” We think it’s hard for a Fed official to be more direct than that. (This ended up being more accurate than perhaps Dudley intended, as a first Fed hike didn’t come until December 2015, partly due to the large fall in oil prices in 2014.)
The repricing of short-term interest rates remains one of the more notable aspects of the taper tantrum. In 2013, markets misread the Fed’s intentions and erroneously extrapolated the desire to “step down” purchases to mean that the Fed was making a bigger change in its accommodative stance.
Encouragingly, this time around markets don’t appear to be making the same mistake. In the US the market pricing of the path of short-term interest rates has actually moved lower this year, even as the change to reinvestment policy has moved closer. And in Europe recent communications about a potential taper of asset purchases has had limited impact, if any, on the pricing of the short-term rate path. This is a key difference from 2013; this time around, central banks have been more successful in maintaining expectations that broad accommodation will remain in place—and should that hold, it will be an important departure from the taper tantrum.
The Negative Correlation Between Bonds and Risk Assets in 2013
A common observation about the taper tantrum in 2013 is that bonds and risk assets both experienced a period of negative returns, thereby upending the negative correlation between the two that is the foundation of many portfolios, including portfolios at Western Asset. This was true, but it was only so for a very limited time period. Specifically, in the month following Bernanke’s comments, the S&P 500 fell by 6% and a broad index of US Treasury bonds fell by 2%. As would be expected in such an environment, corporate bonds performed poorly, both due to the rate exposure and due to the credit component, and the high-yield index lost a total of 5% over the same period.
However, this one-month perspective is almost surely too narrow. If we broaden the horizon to include the six months following Bernanke’s comments, the picture looks different: S&P 500 +8%, US Treasury Index -1.5% and high-yield bonds flat. And if we broaden the horizon a step further to include all of 2013, then the correlations again look normal, insofar as bonds and risk assets were negatively correlated: S&P +25%, US Treasury Index -2.5% and high-yield +6.7%.
Which time horizon is the correct one? While those concerned about the possibility of positive correlations working against a portfolio that has both bonds and risk assets can point to the month following Bernanke’s comments as proof that it can happen, focusing on this one month misses the bigger picture: that 2013 was a standout year for risk assets, and given that, it is not too surprising that US Treasuries (USTs) had modestly negative total returns. When assessed in the proper perspective, such a combination of price moves is neither puzzling nor alarming.
Emerging markets are an exception; they did not recover following the initial sell-off and had a negative year in 2013. However, as explained in the next section, emerging markets are a particular case that had more to do with the starting point than the impact of Fed moves.
The Starting Point Matters: Emerging Markets
Emerging markets were especially hard hit during the taper tantrum, with a broad index of EM bonds returning -10% in 2013. Some commentators have focused on the sensitivity of EM assets to Fed policy as the primary driver of the underperformance. Indeed, 2013 was not the first time that moves from the Fed coincided with dislocations in EMs; other examples include the 1994 Tequila crisis in Mexico and the 1997-1998 balance of payments crisis in Thailand, South Korea and Russia, each of which came during a period of Fed tightening.
While we can’t rule out that EM assets’ sensitivity to Fed policy played a role, it is unlikely that this was the main culprit for EM underperformance in 2013. A more likely explanation, in our view, is that EM assets were starting from a point of rich valuations after a period of exceptionally strong growth and asset inflows. Admittedly, these types of conclusions are always easier to make in hindsight.
Emerging market growth was strong following the 2008 financial crisis, and was especially strong when compared with the more tepid growth in advanced economies. During the five-year period leading up to the taper tantrum, EM growth outpaced advanced economy growth by 5% (5.5% annual rate for EMs versus 0.5% for advanced economies). The outperformance during that period was well in excess of what had been seen in the previous 20 years (EMs outperforming by 1.5%) or in the subsequent five years (EMs outperforming by 2.7%). The reasons for the growth outperformance are beyond the scope of this note, but suffice it to say there were a number of contributors, including a significant Chinese credit stimulus in 2010-2012.
Unsurprisingly, robust growth in EMs coincided with strong inflows and in turn led to an appreciation in asset values. From 2008 to the start of the taper tantrum, EM currencies outperformed the US dollar by more than 15%. Moreover, yields on EM sovereign debt remained quite low, and at the start of the taper tantrum the spread to UST yields was virtually unchanged from the start of 2008, an impressive result considering that spreads in other credit sectors were still quite elevated at that time (Exhibit 2).
This was the starting point—low yields and currencies that had appreciated significantly—when the dynamics shifted in 2013. Most importantly, growth downshifted and the advantage of EMs deteriorated, as there was a slight acceleration in advanced markets at the same time. This was compounded by a fall in commodity prices, a barrage of negative political developments (especially in Russia and Brazil), and increasing inflation in certain cases. The response of EM assets reflected all of these developments: currencies underperformed the US dollar (down more than 20% from the high in May 2013 to the low in early 2016) and yields increased (the spread to USTs increased by more than 200 basis points). After an extended period of underperformance, EMs have had a better year so far in 2017, but currencies still remain low and yields continue to be elevated relative to where they were at the start of the taper tantrum.
This brings us to today and the question of whether changes in central bank balance sheets will again be problematic for EMs. Perhaps they will, but in our view this is only a small part of the story. Much more important will be the starting point for these assets. Unlike in 2013, EMs are entering the current period of central bank adjustments with valuations that are still attractive relative to recent history. Similarly, the growth slowdown has already happened in EMs and it would be quite surprising to see another shift in relative growth of similar magnitude. These fundamentals of value and growth are likely to be bigger drivers of EM asset prices than anything coming from the Fed or the ECB.
Central Bank Balance Sheets, Inflation and Growth
Perhaps the most important lesson from the taper tantrum is that, in the current environment, the fundamentals of growth and inflation have more influence on bond yields than central bank balance sheets. In 2013 the inflation rate was running somewhat below the Fed’s 2% target. However, the Fed was forecasting firming inflation as improvements in labor markets and above potential growth were expected to lead to tighter input markets. Inflation markets appeared to confirm the Fed’s view, as the inflation rate priced into medium-term inflation breakevens were above the 2% target, even though inflation breakevens for the very near term were well below it. This is an important point; the adjustment in balance sheets in 2013 was in anticipation of future inflation, rather than in response to realized inflation.
This optimistic inflation forecast was well off the mark. The surprisingly low inflation prints in the US this year have only been the latest in a series of disappointments. The collapse in oil prices that started in mid-2014, the failure of wages to pick up even as the unemployment rate has fallen, and the very weak global-inflation dynamics (see our recent paper on Global Inflation Rates) have all contributed to the string of below-forecast inflation prints. Due to all of these factors, core inflation has underperformed the Fed’s 2013 forecast by a total of 0.3% at an annual rate. The growth outcomes have been similar, with real GDP growth underperforming the Fed’s forecast by an annual rate of 0.5% over the same period.
So, a number of years later we are now in the position to answer the following question. Which matters more in the current environment: the Fed adjusting its balance sheet policies or the path of growth and inflation? The obvious answer is that the path of growth and inflation are much more important. Yields on 10-year USTs are lower today than they were at the peak of the taper tantrum, even though over the intervening years the Fed has not only adjusted its balance sheet policies but also hiked rates four times. Even if an adjustment of central bank balance sheet policies leads to some turbulence, over the longer term the fundamentals will likely reassert themselves and bond yields will reflect growth and inflation.
The lessons from the taper tantrum are important and relevant for bond investors today. Rather than providing a playbook for how the upcoming period will unfold in markets, however, these lessons instead highlight the ways in which the circumstances today are unlike those of 2013. Central banks have been clear that adjustments to balance sheet policies do not signal a shift in the broader accommodative stance, and so far have been effective at communicating that message. The negative correlation across bonds and risk assets is likely to hold, much like it did over the course of 2013, even though it broke down for a month in the middle of that year. The starting point matters for asset prices, and the starting point for emerging markets today is notably better than it was in 2013. And, finally, over the longer term, markets will respond to inflation and growth, neither of which is likely to be impacted all that much by upcoming adjustments to central bank balance sheets.