What Keeps Goldman Up At Night

Tyler Durden's picture

If one listens to Goldman's chief economist Jan Hatzius these days, it is all roses for the global economy in 2014... much like it was for Goldman at the end of 2010, a case of optimism which went stupendously wrong. Goldman's Dominic Wilson admits as much in a brand new note in which he says, "Our economic and market views for 2014 are quite upbeat." However, unlike the blind faith Goldman had in a recovery that was promptly dashed, this time it is hedging, and as a result has just released the following not titled "Where we worry: Risks to our outlook", where Wilson notes: "After significant equity gains in 2013 and with more of a consensus that US growth will improve, it is important to think about the risks to that view. There are two main ways in which our market outlook could be wrong. The first is that our economic forecasts could be wrong. The second is that our economic forecasts could be right but our view of the market implications of those forecasts could be wrong. We highlight five key risks on each front here."

In short: these are the ten things that keep Goldman up at night: five economic risks, and five market view risks. To summarize:

Five risks to our economic forecast


Our central forecasts look for accelerating growth in the DM economies, against the backdrop of continued low inflation and low policy rates. The main risks to this view are: a reduction in fiscal drag is less of a plus than we expect; deleveraging obstacles continue to weigh on private demand; less effective spare capacity leads to earlier  wage/inflation pressure; Euro area risks resurface; and China financial/credit concerns becomes critical.


Five risks to our market view


Five key risks may affect the mapping of our macro views into the market forecast: long-dated real yields rise more sharply; markets doubt G4 commitment to easy policy in the face of better growth; low risk premia create valuation challenges; margins compress more rapidly as wage share recovers; and EM assets benefit more from the DM recovery or suffer more from local imbalances.

And some more in depth:

Our forecast for 2014 is a relatively upbeat one. At the core of the outlook is a shift in US GDP growth to a sustained period of above-trend growth, alongside improved growth prospects in the developed economies in general. At the same time, owing to substantial slack in the major developed economies, we expect inflation to remain benign and G4 monetary policy to remain relatively easy. Although improving growth is likely to put upward pressure on bond yields, we think the combination of better growth and still-low yield levels remains supportive for risky assets in the developed markets. By contrast, we still see the emerging world as more constrained by tighter capacity and domestic imbalances and more vulnerable to higher global yields.

Although the mood in markets has turned more optimistic, we still encounter nervousness about the capacity for the outlook to remain positive. Much of that anxiety appears to reflect scepticism that after a strong year for stocks in 2013 – and a prolonged recovery period that has lasted nearly five years now – further gains are possible. We think this view underestimates the depth of the hole out of which the major economies and markets are climbing. But there are more concrete challenges to our forecast that go beyond the longevity of the recent rally, and that we worry about and debate ourselves. We look at some of the main issues here.

There are two main ways in which our market outlook could be wrong. The first is that our economic forecasts could be wrong. The second is that our economic forecasts could be right but our view of the market implications of those forecasts could be wrong. We highlight five key risks on each front, although in practice it is hard to separate the two categories so neatly and, if we are wrong in one area, it may increase the chance of being wrong in others.

The biggest issues revolve around two main areas. The first area is the supply side of the global economy and whether better demand growth will increase focus on capacity constraints and put upward pressure on bond yields. Arguably, the single greatest challenge to our market views would come from an environment in which wage and inflation pressure in the US and other developed markets came earlier than we expected, prompting a combination of higher inflation, greater margin erosion, earlier tightening and higher volatility than in our central case. The second area is policy risk, where the key challenge is the exit from extraordinary monetary stimulus, but where fiscal risks and the constraints on EM policymakers are also still critical. Those challenges are heightened by the difficulties of communicating with the market in the unfamiliar terrain of forward guidance and asset purchases. In laying out our Top Market Themes, we have focused on the prospects for hedging against the risk of a sharper rise in bond yields and concerns about premature monetary tightening. The risks below set out a broader range of areas where hedges could be worth exploring.

Five risks to our economic forecast

Our central forecasts look for accelerating growth in the DM economies, against the backdrop of continued low inflation and low policy rates. The shift in the US to sustained above-trend growth, in particular, has been the backbone of our economic and market views for some time. Any forecaster knows that there are many ways to be wrong, so we are not attempting to be comprehensive. But the five risks below would have a material impact on the way we see the economic backdrop. It is hard to balance how likely each risk is with how problematic they would be for our views, so the order should not be interpreted as a ranking:

1. Reduction in fiscal drag is less of a plus than we expect

The core of the story on why we expect US (and DM) growth to accelerate in 2014 is set out in Exhibit 1. The chart shows our estimates of the private- and public-sector impulses for growth in the US. On that basis, we see private-sector healing having already led to an improving impulse but this was masked by significant drag from the public sector in 2013. As that fiscal drag alleviates in 2014, alongside a still-positive private-sector story, we should see overall GDP growth pick up. A key risk is that the alleviation of fiscal drag adds less to growth than we expect. That could happen for one of two reasons. First, underlying fiscal drag in 2013 may turn out to have been smaller than we estimated. If so, this would mean that the private sector has added less to growth and, as the fiscal drag alleviated, any acceleration would be more modest. Second, our assumptions about current fiscal plans may turn out to be too optimistic. We have already highlighted the prospective drag in Japan, where the consumption tax on current plans is set to deliver a sizeable negative fiscal impulse. There is also still some risk that fiscal negotiations in the US break down again in the spring, delivering a fresh fiscal shock, although we think those risks are fairly low.

The first of these may pose the more important risk. There is room for uncertainty here, because fiscal drag is not directly observed. We do observe the shift in government tax and spending plans (and even here true ‘shifts’ are complicated to tease out of the data, in part because they also depend on estimates of spare capacity). To convert those into estimates of drag, we need to assume multipliers into broader activity. The work we (and others, such as the IMF) have done suggests that those multipliers have been quite large given the constraints on monetary policy and credit creation. But it is still possible that we have overestimated the drag from the public sector. The fact that US GDP growth accelerated from 2013H1 to 2013H2, as have measures like our Current Activity Indicator, is comforting for the story that fiscal pressures early in 2013 were a large drag. But it is not yet definitive evidence.

2. Deleveraging obstacles continue to weigh on private demand

The other half of the story of improving growth is that the US private-sector impulse remains robust. As discussed above, the flipside of a smaller estimate of fiscal drag would be that the private sector has not been adding as much as we think. The simplest version of that risk may be that the post-bust headwinds and the pattern of slow recoveries that follow financial crises continue to prevent a self-sustaining pick-up in private demand. US capital spending in particular has been lower than our forecasts. While we expect it to pick up as consumer demand growth improves, there is some risk that excess capacity and uncertainty about demand continue to weigh. In the Euro area, similar risks exist but pressures from ongoing bank deleveraging (lending to the private sector is still contracting) constitute an important additional headwind. The impact of new financial regulations and, in some places, fresh macro-prudential measures could also have a larger growth impact than we anticipate.

In assessing the broad risks to US private-sector demand, the housing market still plays a key role. Directly or indirectly, the housing recovery has been responsible for about a third of US growth over the last year (Exhibit 2). Following the rise in mortgage yields in the summer, we saw a patch of weakness in housing data that has now begun to reverse. Our assumption is that the secondary housing market will continue to improve, although at a more modest rate than last year, and that the scope for new building remains high. But the biggest risk to that view is probably a fresh tightening in financial conditions, which have reversed all of the tapering shock and are back at very easy levels, and mortgage rates in particular. Away from the US, a number of housing markets (Canada, Sweden, Switzerland and even the UK) have also shown significant gains on the back of low interest rates and could prove vulnerable to sharper reversals, rising rates or macro-prudential measures.

3. Less effective spare capacity leads to earlier wage/inflation pressure

As growth accelerates above trend for the first time in the recovery, the supply side of the US and other DM economies is likely to become a more important issue. In particular, our forecasts assume that there is plenty of effective spare capacity in the G4 economies, and that this will keep wage and price inflation well-anchored despite better growth. Underlying this view is that the financial crisis has not led to large permanent losses in potential output and that, while the NAIRU (the lowest unemployment rate consistent with stable inflation) has probably increased, it is still well below current levels. The lack of significant wage inflation suggests that those who argued in recent years that the US NAIRU may now be as high as 7%-7.5% have been wrong. But the question of where capacity pressures begin to bite is always highly uncertain and is likely to be put to a tougher test as growth picks up. The US unemployment rate has also consistently fallen faster than our forecasts anticipated, including over the last 12 months, as the participation rate continues to fall (Exhibit 3).

Surprises have not yet made a difference to the inflation picture and we still think there is plenty of slack. But systematic forecast errors naturally make us less confident. As the unemployment rate itself drops (on our forecast, it ends 2014 at 6.3%) the debate over the drivers of the US participation rate will sharpen. Outside the US, the UK is likely to see similarly intense focus on the issue of slack and labour market tightening. The unemployment rate has fallen more rapidly there too, on the back of the recent acceleration in growth, and we expect it to decline further. In the Euro area as a whole, with slower growth and higher unemployment, we see much smaller risks that capacity tightness will become an issue this year, although we do forecast a significant increase in those pressures in Germany in 2014. Even if we are right that there is plenty of slack in the developed world, the market may worry about these risks before it is convinced. And if growth is more rapid than we forecast – a possibility in the US and UK, at least – spare capacity would be eroded faster, even if our view of the starting point is correct. Our benign view of supply constraints also extends to commodity markets, where we envisage steady oil prices and falls in many other commodity prices. If that view is wrong – or if geopolitical issues disrupt supplies – that would also worsen the broader risk picture.

4. Euro area risks resurface

Although we expect the strongest impulse in 2014 to come from the US, we forecast improving Euro area growth. That improvement – and an expectation that Euro area system-wide risk will remain well-contained again in 2014 – is an important support for our benign view of the outlook. The Euro area starts 2014 on a more stable footing than it has been on for some time. Financial conditions in the periphery have continued to ease, growth has picked up somewhat since the recession in early 2013 and the Euro area should see some of the fiscal relief that is helping the US outlook. Much of the improvement can be traced back to the ECB’s commitment to “do whatever it takes” and the consequent mitigation of the risk of self-fulfilling liquidity crises. But the other economic risks considered above may apply in the Euro area in particular. And, as Huw Pill described late last year (see European Economics Analyst: 13/45 – ‘Euro area adjustment: Past, present and future’, December 19, 3013), there has been patchy progress on the larger adjustments the Euro area has needed to make.

That process, however, is incomplete (Exhibit 4). Even with the improvement in GDP growth that we forecast and the compression in sovereign spreads that we have seen, low nominal GDP (and domestic demand) growth still makes the task of restoring fiscal sustainability an uphill battle. And it serves as a reminder that neither the fiscal nor the real exchange rate adjustments that are necessary to make us confident about a stable longer-term outlook for the Euro area are yet assured. Our view is that with growth improving, concern over these issues is more likely to fall than rise. But a sharper slowing in global growth or disappointments in the local recovery, particularly in Italy, could put these issues more quickly on the agenda. Further falls in Euro area inflation may also reinforce concerns about both the fiscal and real exchange rate adjustments, particularly given worries that the ECB lacks the capacity or will to respond aggressively to that threat.

5. China financial/credit concerns become critical

Our forecast for China is for stable growth in 2014 as the global recovery helps exports, offset by a modest ongoing tightening on the domestic side. Given low expectations, we think that profile, while unimpressive, could still be modestly reassuring. But despite that relatively benign central view, we still see significant risks from the credit imbalances that have built up, an issue that we focused on in detail last year (see our Portfolio Strategists’ report entitled ‘The China credit conundrum: Risks, paths and implications’, July 31, 2013). Nominal credit growth (measured by total social financing) is still running well above nominal GDP growth (Exhibit 5). Moreover, while our forecasts implicitly envisage a narrowing of that gap this year, we do not expect it to reverse, which is what is ultimately needed to prevent debt-to-GDP ratios from rising.

The process of leaning against financial imbalances through ongoing tightening in financial conditions and increased reliance on market interest rates, while important for restoring sustainable growth, is also inherently risky. Over the last few weeks, we have seen renewed spikes in Chinese interbank rates, which have weighed on financial markets. The risk is that tightening proceeds too rapidly sparks an unexpected deterioration in credit conditions. But this is a symptom of the broader challenges of deflating the credit bubble slowly while maintaining steady rates of growth.

Five risks to our market view

The basic market outlook that we forecast is one in which equities and bond yields can continue to rise together and in which DM assets continue to offer better risk-reward than their EM counterparts. Clearly, if our economic views are wrong, that entails risk to our market views. But even if the economic landscape is broadly as we expect, the market implications of that outlook may be different to those we forecast. We see five big risks that may affect the mapping of our macro views into the market forecast:

1. Long-dated real yields rise more sharply

We forecast that real bond yields in the US (and G4) will rise modestly in 2014, at a pace close to the forwards. Given the importance of the valuation gap between bonds and equities for our story, we have highlighted the risk that this gap closes more through higher real yields than through higher equity markets. We have argued that this is probably the most likely (although not necessarily the most damaging) risk to our broad market views and, as a result, our Top Trade recommendations have been designed in part to protect against it (through short rate and long USD positions). There are good reasons to expect the rise in yields to be relatively moderate. Back-dated yields have moved much closer to normal levels than they were a year ago and our measures of G3 bond valuations are no longer ‘rich’. Inflation remains benign and major central banks still look committed to a long period of unusually low policy rates. And with Fed tapering already begun, the market has incorporated the expectation that asset purchases will end this year. But the term premium in bonds is still lower than during much of the pre-2007 period. And models of the term premium and fund rate paths that we used last March to argue that there was upside risk to yields suggest that some upside risk remains (Exhibit 6).

Exhibit 6 shows a range of yield estimates generated from simple term premium models. While these do not point to large upside risks in 2014, they do imply increased risk beyond that point. These models are based on our own benign view of the Fed funds path and inflation, so if those assumptions come under challenge, so could our yield view. We have argued that rising yields are consistent with positive equity returns as long as improving growth is driving both higher, so the driver of higher yields matters. That is a key reason why we have been mostly relaxed about the impact of higher bond yields so far. But there are two key caveats. The first is that periods where yields rise rapidly – even with strong growth – tend to hurt equities, as during the taper tantrum. Temporary periods of that kind are likely even in our central forecast. The second is that if yields rise because of expectations of tighter monetary policy, this is more unambiguously negative for equity markets. On the rates strategy front, it is the pricing of rates over the next 3-5 years where we think the risk premium looks unusually low. That takes us to our next risk.

2. Markets doubt G4 commitment to easy policy in the face of better growth

Our forecast is that although tapering will likely be completed this year, the Fed will not raise the Funds rate until early 2016. The Fed itself has worked relatively hard to try to decouple the decision to taper from the decision to hike. And while the market gave little credit to that view during the ‘taper tantrum’ last summer, expectations of the path for US policy rates fell back significantly after the non-taper in September and the nomination of incoming Fed Chair Yellen, despite rising bond yields. Despite our expectations of a friendly Fed policy stance, stronger growth is likely to make it harder to stop markets worrying about an earlier rate hike. Exhibit 7 shows that the volatility in forward views of the US policy rate has tended to be higher during periods of positive data surprises. The same dynamic is true in the UK. Over the last few weeks, as the data have improved we have already seen a fresh rise in shorter-dated yields, so anxieties here have not been calmed completely.

If growth moves decisively above 3% and the unemployment rate continues to fall steadily, the market is likely to worry more about the Fed’s resolve unless guidance is very clear. The decision not to lower the unemployment rate threshold in December, but to rely instead on a less precise description of when rates are likely to rise (“well past the time that the unemployment rate declines below 6-1/2 percent”), may have increased that risk. We expect both the UK and US to breach the formal unemployment thresholds this year. While these kinds of communication challenges will be an issue even if central banks remain committed to their current strategy, policymakers themselves may begin to sound less committed to prolonged easy policy as the recovery becomes more visible. That is clearly possible in the UK and US. But there is also a risk that the Bank of Japan may sound less committed to sustained asset purchases as inflation picks up. A premature focus on their exit strategy would pose a risk to bullish Japanese equity and bearish JPY views.

3. Low risk premia create valuation challenges

Beyond the challenges from rates markets, concerns about valuations – not just in bonds and credit but in equities too – have clearly intensified. Some of those concerns we think are really worries about the longevity of the rally rather than the levels of markets. But as the expansion and market rally have progressed, risk premia have clearly fallen and equity valuations in particular are cited more often as a constraint. On most absolute measures of valuation (P/E multiples, price-to-book, cyclically-adjusted P/E ratios), equities look modestly expensive relative to history, although they are well below the levels that have signalled significant obstacles to gains in the past. On relative measures of valuation compared with bonds or credit – the equity risk premium or equity credit premium – equities still look cheap relative to history, although less cheap than last year. So a key question (see the next risk) is whether real yields remain lower than normal and whether relative or absolute valuations are in the driver’s seat. Measures of the equity risk premium are derived rather than observed and we are conscious that more direct measures of risk premium – the VIX and credit spreads – are starting 2014 at their lowest levels since 2005-07 (Exhibit 8). So the tailwind from risk compression is less powerful than it has been.

Our view is that that relative story will remain supportive for equities, although we agree that return prospects are lower than they were coming into 2013. We think the macro environment remains one that supports low volatility and tight credit spreads. We have shown in the past that the level of the VIX is broadly related to the change in the US unemployment rate, and is also generally lower when growth is higher. As a result, volatility tends to decline during recovery periods and pick up only as the expansion matures. With another year of steady declines in the unemployment rate likely in 2014, this argues for another year of lower US volatility. But this relationship also highlights that the risk of an earlier US labour market tightening than we expect could bring forward the point at which equity volatility begins to rise. A rising volatility environment need not be inconsistent with rising equity markets, as the late 1990s powerfully demonstrated. But it would increase the risks.

4. Margins compress more rapidly as wage share recovers

One of the key supports for US equities over the last few years has been the high level of corporate margins. The US profit share has remained at historically high levels. Our expectation is that it will remain there in 2014. As described in the final US Economics Analyst of 2013, we expect price inflation to be a touch higher than unit labour cost inflation, which has historically allowed firms to maintain or improve margins.

Once again, this depends on the assumption that wage growth will remain benign and implicitly that the labour market has plenty of slack. Exhibit 9 shows that, in general, there is a strong tendency for the profit share to decline when the unemployment rate falls below the NAIRU. Given our view that we are still well above that rate and likely to remain there through 2014, it is logical to expect little margin pressure in this environment. But this means that the issue of labour market slack is relevant not just to the inflation and policy outlook, but also to the margin story. A more rapid tightening in the US labour market represents the most significant risk to record-high margins.

5. EM assets benefit more from DM recovery or suffer more from local imbalances

We have had a structurally cautious view of the EM outlook over the last year or two. As US growth and yields have begun to pick up, we have argued that this puts some of the imbalances that have built up in EM more squarely into focus. And although we expect growth to pick up slightly in EM and modest positive returns from equities in 2014, our outlook – both for EM economies and markets – is still more cautious than for the DM economies. There are risks on both sides of the EM outlook. On the downside, we could easily see greater pressure than we forecast on a range of EM economies and markets from the ongoing rise in US yields and the rebalancing of their own economies. And EM equities and currencies have generally struggled again on the back of these dynamics as 2013 has drawn to a close. A weaker growth outlook in China would be an additional negative force for many EM assets. And elections in a number of countries (Turkey, Brazil, India) may also increase political risk. Given our relative caution, this would to some extent be a reinforcement of the basic picture we expect. As a result, the main risk to our own forecasts – and perhaps the markets too – would be that EM assets do better than their DM counterparts, reversing some of the recent underperformance.

On that front, we see two main risks. The first is that we may be underestimating the impact of an improving DM demand picture on EM economic and market outcomes.
Weakness in the major advanced economies has damaged EM export growth and increased reliance on domestic demand, deteriorating current account positions in the process (Exhibit 10). An improving external environment could help to alleviate both pressures. Our forecasts capture some of that. And we agree that countries, sectors and assets that are most exposed to the DM demand cycle are likely to outperform within EM. But implicitly we are assuming that this will be insufficient for EM assets to benefit from a DM cyclical pick-up in the way that they did through much of the last expansion. We think that assumption is sensible, but it has not been tested and could prove too negative. The second risk is that many of the relationships that have held between DM and EM assets in recent years, particularly in rates and credit, show that EM assets are now trading at a notable discount. Our views imply that some of the breakdown in these relationships is likely to be persistent. We think the regime of the last 5-10 years is likely to continue to prove a poor guide to the next year or two in EM. But, once again, the risk is that the world has changed less than we think and that, with investors now less exposed to EM, the increased risk premium is sufficient to draw money back in.

The key risks: Lack of slack, policy missteps, no trend for a friend

Pulling the various pieces together, it becomes clear that the question of how much ‘effective slack’ there is in the global economy – and the US labour market in particular – is likely to become a much more critical part of the outlook as growth moves above trend and runs through several of the ten risks we have considered. Above-trend growth may also provide a tougher test of our view that there is still significant room for the expansion to continue before it hits supply constraints. If we are wrong on that assumption, not only will this shift the market’s view on how long policy rates will stay near zero in the major markets and tilt the risks towards a faster rise in long-term bond yields. It may also lead to more rapid margin compression and an earlier rise in equity volatility than we currently expect. Our confidence in our view here is relatively high. But our analysis of the risks highlights the importance of that assumption and the benefits of protecting against it. We have argued for hedging against some of these risks in our Top Themes and Trades, emphasising opportunities to be short the US 5-year segment and long the USD against some commodity and EM currencies and the JPY.

The chance of policy missteps also runs through many of the risks we worry about. A better growth picture will heighten focus even further on monetary policy exit. Even if policymakers stay the course with respect to their commitment to easy policy, they may struggle to convince markets. And structural risks in China, the broader EM and the Euro area present additional problems that have not yet been convincingly resolved.

Most of the risks above focus on circumstances in which the market environment turns much less benign. Understandably, these risks are our biggest source of concern. But a more likely source of challenges that we also worry about for 2014 may not be that markets are bad, but that they are boring. With the market broadly in agreement that the growth picture is improving, and with risk premia lower than before, there is a clear risk that we simply enter a period of low expected returns, tracking along forward paths in the major assets. That would not be a disastrous outcome for investors. But it would make the Sharpe ratio on many assets low, both for long and short positions. And navigating the inevitable fluctuations around a subdued trend could make it easy to make tactical missteps that eat into already low returns. We think the picture is likely to be brighter than that. But this – as much as a major disruption – is one of our biggest worries.