Investment Insights

Blow-out, or blow-up: Why risk still matters (long version)

A bear market on bonds? Not yet, but if the dials on the inflation machine keep lighting up, then that may quickly change.


Simon Doyle

Simon Doyle

Head of Fixed Income & Multi-Asset


The first and almost final draft of this paper was written in late January, but a few days in markets can make a world of difference. While I’ve incorporated the early February acceleration in volatility into the text, the essential conclusions are unchanged and the focus of the paper remains on the medium-term outlook, as who knows where we’ll end up in the short term.


Investors with a cautious view of the world ended 2017 with egg on their faces. Equity markets reached new highs, credit spreads narrowed significantly and investors generally were rewarded for taking risk. The more risk taken, the better the returns. Geopolitical risk did not derail asset markets, nor did increasingly demanding valuations. Instead, relatively synchronised global growth, rising profits and low inflation have allowed central banks to continue to tread gently. The party rolled on.

We did not expect 2017 to go as well as it did. We were among the nay-sayers arguing that relatively full valuations made markets vulnerable and while we did not expect that this alone would topple markets, it did mean future returns would be modest, at best. While we may ultimately be right, 2017 was clearly not the year.

There’s a good chance that if you’re reading this note you’re probably fully aware of the challenges involved in forecasting. Despite best endeavours it is an inherently difficult and potentially dangerous task given the interactions between politics, policy making and human behaviour. If the truth be known, being right in the short run is probably just as much a function of luck as it is skill and we shouldn’t delude ourselves into thinking otherwise. The lack of predictability in the short run is after all why we build portfolios, using portfolio construction and diversification to counterbalance the uncertainty in any set of forecasts. Forecasting helps in assessing where the risks are but is no guarantee that markets will fall in line.

So with our forecasting credibility heavily caveated, we outline in this note what we see as the key issues for 2018, how this could play out in financial markets, where the risks lie, and how we are approaching these challenges from a portfolio construction perspective.

In essence and notwithstanding early February falls in risk assets, there are two critical differences between today and 12 months ago. The first is that valuations in risk assets are materially more demanding, particularly at the end of January. Equity multiples are higher and credit spreads narrower. The second is that the preconditions for rising core inflation have continued to build. Tighter labour markets, growing anecdotal evidence of emerging wage pressure (especially in the US) and evidence that output gaps have largely closed is consistent with inflation building this year with the cycle more progressed. This is a critical issue. Central bank actions (or inaction) have become an essential underpinning of market behaviour and it is on inflation that monetary policy will pivot. If of course inflation doesn’t rise from here it likely means either that structural disinflation forces have overridden the cyclical story (long live the bull market), OR, that recession is unfolding. The latter would also be damaging for markets.

The difficulty though is timing, mainly because the macro-economic and policy environment is still broadly favourable. However, as recent volatility demonstrates, markets are vulnerable. But, as history tells us, that alone is not sufficient. Equities were expensive in 1998 and 1999 and didn’t fall sharply until 2000. Likewise credit spent 3 years in the mid 2000’s at lower spreads than they are at today before unravelling through the GFC. For equities in particular, it is entirely possible that in the near term further gains accrue as markets “blow-out” in the absence of an economic or policy shock. What could this “shock be”? As noted above, rising inflation will be the most probable culprit, and evidence is accruing that the risks here are building. The more material challenge for markets would be recession and we do not think this is likely on a 12 month view.

Sovereign bond markets have had a shaky start to 2018 after trading in a pretty narrow range in 2017. We have argued for some time (without much reward) that there is a mismatch between the level of yields prevailing in key sovereign markets (especially the US) and economic conditions and were unsustainable in the medium term and that remains our base case. While we are encouraged in this view by the recent rise in yields, in the absence of hard evidence of rising core inflation, we will curb our enthusiasm in calling a start to a “bear” market in bonds as it overstates the impact on overall bond returns, albeit we do expect sovereign yields to continue to rise from here. The bigger risk from rising yields is on financially leveraged assets (like A-REITS and infrastructure) that have (at least until recently) seen significant price appreciation directly related to lower yields.  

How are we dealing with this in portfolios?

Fortunately our job is not to accurately forecast market returns over the next 12 months, but to invest client money to achieve agreed performance objectives amid inherent uncertainty. Forecasting is a means to an end, but not the end itself. Investment and portfolio construction implications are more important. In this context we believe the implications of the current environment are as follows:

  • The valuations embedded in most assets make low nominal and real returns at an asset class level over the medium term a near certainty.
  • As we’ve recently been reminded, the path to lower returns won’t be smooth. Low volatility is not the norm. While 2018 might again defy gravity, this will simply make the outlook beyond that more challenging.

More specifically:

  • Our well documented concerns about the mismatch between volatility and risk in asset markets has been validated in early February with sharp falls in risk assets and rising market volatility;
  • That said, we don’t expect a major bear market in risk assets to start just yet but current valuations and the asymmetric risk in bonds and credit suggests heightened vulnerability;
  • Returns from risk assets (equities and credit) will moderate as valuation constraints exert themselves;
  • We are seeing a spike in volatility and expect heightened volatility to remain, largely on the back of rising inflation and growing uncertainty about how Central Banks will respond;
  • We are relatively “cashed-up” and focussed on high quality assets with reasonable liquidity. All else being equal, we expect to be tactical buyers of risk assets on weakness;
  • Investments and strategies that perform well in moderately rising inflation environments are favoured and we continue to incorporate these into our portfolios. These include: cash; inflation linked bonds; break-even inflation positions; and the avoidance of bond sensitive sectors (like A-REIT’s);
  • With broad markets likely to deliver moderate returns (at best) we believe cross market strategies will be important sources of return. In this we include currencies where volatility is more “normal” and both strategic and tactical mispricing evident (eg. GBP and USD v AUD);
  • The risks around this outlook we’d simplistically describe as either “blow-out” or “blow-up”.  The critical question is which is the most important. For us, a “blow-up” takes precedence as the minimisation of drawdown risk is crucial to the achievement of our objectives across most of our client portfolios.
  • While we may again leave some return on the table by not adopting a more aggressive near term stance, we are not uncomfortable with taking our chance here given that the medium term risks favour the “blow-up” scenario.

The building blocks of our view

1. Valuations – a clear headwind

While this note is focussed on the 2018 outlook, that is not independent of the medium to longer term backdrop. Medium-to-long run valuation metrics summarise the valuation challenge. While the Shiller PE framework has its critics (and criticisms[1]), it does provide a relatively independent and agnostic starting point, has a long and consistent history, and has a strong relationship with prospective returns over the medium to longer term. The end January Shiller PE of 34 x is at its highest level since later phases of the late 1990’s tech bubble and implies returns of around 0% pa in real terms over the next decade.

Outside the US, the extent to which structural valuations are stretched is less evident but nonetheless has similar - although less extreme implications for returns.

It is also worth decomposing recent market performance.  While for the US earnings growth has contributed to market performance, a rerating of the market to higher multiples has been a more significant contributor. This rerating is potentially justifiable should the strong growth, low inflation, low yield environment endure and companies deliver on optimistic assumptions about future earnings growth in effective perpetuity. However as we will highlight later this seems a particularly optimistic and unrealistic assumption.

For bonds, the story is relatively bleak and largely a function of low starting yields. Furthermore, if we are right on inflation then yields are likely to move higher reducing near term returns to bond investors further.

While there is considerable hype about a potential bond bear market, I’d offer a more considered view. Clearly we need to place current cyclical developments into structural context. On the latter, broader deflationary forces from factors such as demographics and technology are still important in setting the longer run trend in inflation. What we are talking about is a cyclical upswing against this backdrop. Furthermore, rising bond yields would lead to very low / negative returns from longer dated bonds in particular, but these will be moderate against the returns from bond sensitive and even risk assets should markets start to reflect appropriate risk premium and policy (or even policy expectations) adjust. We’d also note that higher yields reset coupons and reinvestment rates over time so bond bear markets tend to be more feared than they should be in reality. Other assets are more vulnerable.

Valuations tell us about future returns and risk with their usefulness increasing over time. Our valuation framework has been well documented but essentially combines long run structural return drivers with cyclical valuation measures, summarized for practicality and convenience in both 10 and 3 year return forecasts[2].  Figure 3 compares historical returns versus our latest forecasts. To be fair, an equivalent table 3 years ago suggested returns would have been materially lower than they have been. The perils of forecasting …

While these are forecasts, and should be treated cautiously (and not too literally), there are several important implications.

Firstly, the next few years are likely to be significantly more challenging than the past. That’s not to say 2018 couldn’t ultimately be a good year, but extended valuations tell us that positive future expectations have already been discounted. It is hard to see how markets generally can sustain past returns. The implications of valuations increase as timeframes get longer. It is worth calling out a couple of key markets here. US equities for example offer on our assumptions low nominal returns (2.5% pa) and likely 0% real returns over the next decade. While this may seem low and outlandish against recent trends it is not without precedent. US equities did similar (or worse) in the decade of 70’s and in the decade of the 2000’s. Our 3 year numbers for the US look even more problematic. Bonds too (and this is a more obvious point to draw given current low starting yields in most markets) will likely deliver similar over the long run. This is particularly problematic for Balanced Fund investors with fixed SAA’s, effectively locking in low returns over the medium term.

Secondly, these numbers imply for many assets (particularly evident in US assets) no prospective risk premium. Investors aren’t getting rewarded over time for taking on additional asset class risk and don’t seem to care. The influence of Central Banks has clearly been at work here. Our forecasts for US equities, high yield debt, corporate bonds and government bonds are essentially the same number. Why take equity risk if you’re not likely to get paid for it?

Thirdly, the most bullish observation we’d make from this is with respect to Australian equities. Australian equities have been a significant underperformer in this upswing and offer materially higher yields and less demanding valuations for investors. Of course in the near term they’ll be correlated to global markets, but over the medium they look a better investment prospect (particularly for Australian taxpaying investors).

2. The Business Cycle is a positive

Even the most bearish market forecasters would likely observe that the current global economic backdrop is quite favourable for markets. GDP growth in most major economies is running at decent rates and the risk of recession in the near term (say next 12 months) is relatively low. Adding to the positive economic backdrop is the still modest trend in core inflation which has allowed central banks globally (and even in the US where rates have started to rise) to tread gently. Furthermore, pro-cyclical fiscal policy (aka Trumps tax cuts) is adding to both confidence and economic growth in the near term. Our own assessment of the cyclical backdrop is consistent with this.

The potential for core inflation to rise though is more likely (and is our base case)[3].  On this point the market is still quite bifurcated. Those not worried about inflation essentially argue that the structural deflationary forces (technology, demographics, and the death of the Phillips curve[4]) will overwhelm any cyclical pressure from diminishing output gaps and tightening labour markets. We accept that these  structural factors are real and exerting pressure on inflation over the long run, but we believe there is mounting evidence of building pressure on core inflation (especially in the key US market) and this will start to exert itself in coming months. Our arguments rest on the idea supported by the Federal Reserve and other academic research suggesting that the Phillips curve is non-linear and that we are approaching the levels of unemployment in the US where wage growth will assert itself (maybe mild by past standards but nonetheless still evident). Anecdotal and hard data support this premise and we expect to see evidence continue to mount as 2018 unfolds. Walmart, known for being a tough employer has recently agreed to raise its starting wage and expand benefits to employees. The New York Times recently reported on the impact of the tightening US labour market, highlighting that in Dane County, Wisconsin where the unemployment rate is just 2%, manufacturers are hiring prison inmates on full wages to work in factories while they serve out their sentences. 

The critical questions will be: how do policymakers respond? ;Are central banks behind the curve? Or, perhaps more importantly, how significant this shift in inflation is to investor perceptions about the prices they have increasingly been prepared to pay for assets.

3. Risks

3.1 The case for blow-out

The above outlines both the vulnerability and medium term risk implied by valuations and the more positive cyclical backdrop.  Experience tells us that bull markets rarely end simply because multiples look stretched. Something generally needs to go wrong to bring the market cycle to an end. Policy mistakes, recession, geo-political events are the typical catalysts here. Valuations generally look fragile and vulnerable long before “the market” responds. The tech boom of the late 1990’s was a great example. It was almost 4 years after Alan Greenspan’s “irrational exuberance” speech that the tech bubble burst. Likewise the warning signs for a bursting of the credit bubble in the mid 2000’s were evident several years before the bubble burst.

There are clearly parallels to the current environment.

Valuation metrics are not absolute and could still become more demanding before they peak. History provides a guide, but not hard and fast rules and just as policy rules have been rewritten in the past decade there’s nothing to stop markets doing the same (who needs compensation for risk anyway – volatility’s dead right!).  The current Shiller PE for example for the US market at 34x is still below the peak in the low 40’s experienced in early 2000 before the tech bubble burst. If it were to peak in say the next 12 months at similar levels to the 2000 peak, the US S&P 500 could rise / rebound significantly from current levels.

Secondly, central banks remain “gun-shy” and so monetary policy normalisation (even in the face of normalising macro conditions) is likely to remain as market friendly as possible. The political pressure to avoid another “taper-tantrum” is no doubt acute even if the Fed’s ability to achieve this has diminished. Thirdly, even if we’re right and inflation does start to emerge, moderate inflation itself is not a bad thing for equities as it allows corporates to expand margins and nominal profits to lift.  

3.2  The case for blow-up

There is a fine line between a “blow-out” and a “blow-up”. Notwithstanding the arguments made above and the early February correction, market valuations aren’t cheap and are supported by relatively utopian assumptions about ongoing growth, low inflation and policy accommodation. In a logical world therefore the catalyst for a “blow-up” lie’s with anything that challenges the fundamental macro and policy support for the market. A growth shock (potentially China), a surprising pick-up in wages / core inflation (especially in the US or Europe) or a policy shock (such as Trade-War) or central banks either failing to respond adequately to rising inflation or indicating a more aggressive path to normalisation would all have the potential to derail investor comfort.

The near euphoric sentiment evident in January was a good precursor to February’s repricing. The bigger risks though are fundamental and rising inflation is a legitimate cause for caution.


  • Our well documented concerns about the mismatch between volatility and risk in asset markets has been validated in early February with sharp falls in risk assets and rising market volatility;
  • This fits with our underlying view that recent returns from markets are unsustainable, propped up by extraordinarily accommodative policy settings which are out of line with economic fundamentals;
  • Our caution in 2017 proved unwarranted – but 2018 has begun to unfold with more caution and some rethinking of recent optimism. Our base case is that 2018 will be a year of more moderate returns and heightened volatility so are encouraged by recent developments even if our positioning for these events was premature;
  • There are both upside and downside risks to this view. While we expect (all else being equal to be tactical buyers of risk assets on weakness), we are closer to the end of this cycle than the start and expect to remain broadly defensive and focussed on capital preservation.

[1] The main criticism at present is that as it uses 10 year trailing earnings, and we haven’t had a recession in that time (trough to peak earnings) that it overstates the valuation multiple of the market calculated in this way.

[2] For equities, long run returns are generated by decomposing returns into three elements: income (Y); growth in income (G); and the effect of changes in valuations (ΔV).  For equities, income is the current dividend yield; growth of income is EPS growth, with forecasts based on real GDP per capita growth plus inflation; and the valuation effect is based on the expected change in the PE ratio over the longer run. The forecast of the terminal PE ratio is determined by using assumptions about the inflation environment and taking into account the tendency of PE to revert to the mean (we proxy this using a 30 year average).  We then adjust this long run return for shorter run valuations which are in turn a function of earnings yield relative to cash, and earnings relative to trend.

[3] "Inflation and markets - where to from here?" White paper by Simon Stevenson - Head of Strategy, October 2017

[4]The relationship between the unemployment rate and wages

Important Information:
Opinions, estimates and projections in this article constitute the current judgement of the author as of the date of this article. They do not necessarily reflect the opinions of Schroder Investment Management Australia Limited, ABN 22 000 443 274, AFS Licence 226473 ("Schroders") or any member of the Schroders Group and are subject to change without notice. In preparing this document, we have relied upon and assumed, without independent verification, the accuracy and completeness of all information available from public sources or which was otherwise reviewed by us. Schroders does not give any warranty as to the accuracy, reliability or completeness of information which is contained in this article. Except insofar as liability under any statute cannot be excluded, Schroders and its directors, employees, consultants or any company in the Schroders Group do not accept any liability (whether arising in contract, in tort or negligence or otherwise) for any error or omission in this article or for any resulting loss or damage (whether direct, indirect, consequential or otherwise) suffered by the recipient of this article or any other person. This document does not contain, and should not be relied on as containing any investment, accounting, legal or tax advice. Schroders may record and monitor telephone calls for security, training and compliance purposes.