WILLIAM DE VIJLDER Blog - in English

Difficult period of the year

Published in Moneytalk (Belgium) on 12 May 2012 

It is May, and that means investors’ thoughts automatically turn to the old adage of ‘sell in May and go away’. I’ve always had some problems accepting this advice because stock markets don’t follow a deterministic pattern. And is this ‘investor’s wisdom’ actually borne out by the figures? The following chart shows average monthly price movements (excluding dividends, therefore) on the S&P index over the past 10, 20, 30 and 40 years, based on the Robert Shiller database.

One thing that stands out is that monthly returns have fluctuated more over the past ten years than previously, whereas the patterns in the averages over the past 20, 30 and 40 years are far more in line with each other. The most recent period, therefore, has been highly atypical. There has also been some value in the ‘sell in May’ adage in the past ten years as the returns turned out negative. Although the picture over a longer period is less pronounced, returns during the summer months have nevertheless been lower than those at other times of the year. In other words, there is certainly a seasonal effect. It is not, however, that May in particular is a difficult month, but rather the whole period from June to the end of October. Literally deciding to sell your investments at the start of May would seem somewhat premature, therefore. Interestingly enough, I heard an analyst on CNBC say last week that this year the ‘sell in May’ advice would not apply until June. This immediately focused attention on the fact that ‘sell in May’ is not a law in itself, and that we need therefore to look at possible catalysts for price corrections. Memories of the past two years are still fresh, with regular fears of a new recession in the USA and another new phase in the euro crisis. Given developments in recent days in Spain and Greece, the situation in the eurozone – not surprisingly – is currently the prime candidate for putting investors’ nerves to the test. In the case of Spain I expect the storm created by the government’s support for the banks to blow over, if only because pressure from Europe, to name just one example, means the government is keen to achieve progress in this area. And achieving this progress is vital if international investors are to be persuaded to trust Spain again. Greece, however, is a source of greater anxiety, with the prospect of new elections (which would be held on 17 June and would de facto be a referendum on membership of the euro) against a background of austerity fatigue and the state’s empty coffers. That means at least another four weeks of uncertainty ahead, and we can only hope that, second time around, the result will allow the country to form a government without too much delay.

A second area investors are focusing on is the USA. There, the picture is clearly an improvement on last year, with no talk of a ‘double dip’. Economic growth is reasonable, while housing market activity seems (finally) to have turned a corner and to be moving in the right direction. Corporate earnings, too, are rising, while margins are good, and the Federal Reserve is continuing to operate very flexible policies and is on standby for QE3 if the stormy weather returns. The big question, however, will come after the Presidential elections: in other words, is the US finally going to do something about its budget deficit without choking off growth?

A third ‘hot topic’ is China, where growth in industrial production is slowing, but where inflation, too, is falling and so creating scope for more a flexible monetary policy via the banks’ reserve requirements. The most likely scenario for the Chinese economy would seem to be a soft landing.

To conclude, the financial markets’ summer is likely to be dominated by developments in Europe and, therefore, by a great deal of uncertainty. Although much of this uncertainty has already been discounted in prices (just look at the price/earnings ratios or dividend yields, which are around three times as high as German ten-year yields!), investors are still waiting for the ‘all clear’ signal. And that means we are set to continue in the ‘risk on/risk off’ environment, with rapid switching between equities and safer assets for quite some time to come.

William De Vijlder

Chief Investment Officer, Strategy & Partners

BNP Paribas Investment Partners

Making sense of RoRo

Published in Les Echos (France) and Le Jeudi (Luxembourg) on 25 April 2012.

After the euphoria early this year thanks to the long-term refinancing operations (LTROs) of the ECB, doubts about the Chinese economy, but above all about the Eurozone, once again gained the upper hand. Not only did the old doom-and-gloom scenarios about the Eurozone re-emerge, but so did the associated, somewhat time-weary, metaphor about ‘the marriage that is ripe for divorce’. As for the latter, let me make one thing entirely clear: that comparison is really unfortunate, because it completely ignores the huge cost of such an operation. What’s more, the comparison is exaggerated, because it underestimates the determination of the political leaders to reach a sustainable solution as well as the scale of the financial apparatus that has already been put in place for this purpose. As for the doom-and-gloom scenarios, in the really dark days, people tend to feel that ‘things can only get worse’, as if we were in an out-of-control train hurtling towards a ravine. Time and again, there are calls for a new LTRO or even unlimited bond purchasing by the ECB (the famous bazooka). Reality is different, however, and days of pessimism alternate with renewed optimism or perhaps simply indifference to the challenges facing the Eurozone. This illustrates the RoRo mentality of the markets (‘risk on/risk off’) and the related extremely short investment horizons (a few days), at least of certain investors. The reaction of others (I fear) is ‘I have had enough’: with them, RoRo leads to a complete ‘switch off’, just like in the old days, before the advent of ‘video on demand’, frustrated viewers would switch off the TV when there was nothing interesting on the box.

This RoRo world also reduces the probability of a new LTRO any time soon. Why should the ECB dip into its stocks of ammunition if markets react like a manic-depressive person? Take Tuesday 18 April, for instance. The Eurostoxx received a powerful impulse after the Spanish auction of short-term paper. So was it such a roaring success? No, though both optimists and pessimists found arguments to justify their views; the former because demand was extremely high compared to supply, the latter because the interest rate was a good deal higher than at the previous auction and because the large demand could be interpreted as a sign that investors were shifting their preference to the short end of the curve at the expense of the long end. The optimists won the day and the stock exchange moved higher. But it could just as easily have gone the other way.

Is it possible to rationalise such a RoRo market environment? Yes, because it illustrates the age-old combination of fear and greed – which is precisely the combination that creates bubbles (a centuries-old recurrent phenomenon) and unbridled panic (which is equally recurrent). Perhaps ‘fear and frustration’ is a more fitting description of the situation today; fear that ‘this crisis can go much further’ and frustration about the negative real interest rates. Interestingly, it is this frustration that adds a bright note to the RoRo world, because it limits the danger of a protracted downward trend: in view of the negative real interest rates, investors end up deciding to take some risk after all. And if the decline were to last too long, then the central bankers would ultimately add fuel to that frustration by printing more money.

William De Vijlder
Chief Investment Officer, Strategy & Partners
BNP Paribas Investment Partners

Fears of US bond bear market are premature

Published in Moneytalk (Belgium) on 23 March 2012 and in Le Jeudi (Luxembourg) on 4 April 2012.

The recent improvement in the US economic data is unmistakable, though Ben Bernanke continues to adopt a cautious tone, evidently wary of counting his chickens before they hatch. This caution also reflects his concern not to alarm investors with any suggestions that the Fed’s first interest rate rise may come much earlier than the promised date of end of 2014. This, however, does not alter the fact that long-term interest rates have recently risen quite considerably. The rise since this year’s low in late January is already 40 basis points, sufficient to spark comments that the bear market in bonds has already started. This question matters for numerous reasons. Not-so-young investors will remember the bond crash of 1994 and that’s enough to make anyone nervous. What’s more, higher US interest rates dampen the investor’s tendency to seek his fortunes (= return) far beyond the borders, which could depress emerging country equities and bonds. The question is also relevant for Europe: there is a strong historical correlation between the movement of US and German long-term interest rates. And, given the weak economy in the Eurozone and the latent nervousness about a resurgence of the crisis atmosphere, an increase in these long-term rates is the last thing we need.

But are fears of a possible increase in US long-term interest rates justified at this stage? For various reasons, the answer is a clear “no”. As long as the Fed makes no moves to hike the policy rate, the current steep yield puts an upper limit on the potential increase in long-term interest rates: the short-term interest rate is about zero, so a person who invests in short-term paper instead of a ten-year bond has an opportunity cost of about 2% (the bond coupon). Short-term investing, therefore, is only rational if the investor thinks that the price loss on the bond will neutralise the higher coupon. Long-term interest rates would have to rise by about 20 basis points in the coming twelve months to break even between staying in cash for the coming year and investing in bonds today (due to the duration effect, the assumed interest rate increase would cause a price loss on the bond that would neutralise the coupon). In addition, in view of the interest rate increase that has already occurred, an additional rise would only take place if the Fed were to make noises pointing to a policy tightening towards the spring of 2013. This seems unlikely. A second reason is that, despite signs of improvement, the economy is still far from strong. Unemployment remains stubbornly high, the housing market is barely bottoming out, and the confidence of SMEs and households is still far below the pre-crisis levels. Important trading partners of the US (China and Europe) have, respectively, a decelerating and a weak economy. The third, and possibly most important, reason is that we must prepare for substantial budget cuts in the US once the presidential elections are behind us. Budget cuts would be positive for the bond market as the slower accumulation of debt (due to the smaller deficits) will mean less nervousness among investors, leading to lower interest rates. Moreover, a relaxed monetary policy could help to compensate the negative growth impact of the budgetary restructuring. So the signs for next year are that the US will adopt the same policy mix of monetary relaxation and budgetary tightening currently applied in the UK under prime minister Cameron.

William De Vijlder
Chief Investment Officer, Strategy & Partners
BNP Paribas Investment Partners

Factor approach to investment brings greater correlation

Published in l’Agefi (Switzerland) on 27 March 2012 and in Beleggers Belangen (Netherlands) on 27 March 2012. 

The stock markets have been extraordinarily buoyant since the start of the year. For January/February, we have to go back to 1991 to find a higher excess return on the MSCI World index compared with US Treasury bills. Many other asset classes also turned in good or excellent performances: corporate bonds, emerging market bonds, emerging market currencies, commodities etc. There was clearly a 'risk-on' environment, supported by liquidity creation (the ECB’s LTRO), and an impression that the economic situation is getting slightly better, with the US economy leading the way. 

The problem is that if ‘everything’ is rising in price, the correlations are high. As we know, there is no such thing as a free lunch: when someone invites you, they always have ulterior motives... For a long time economists regarded portfolio diversification as the only exception. The idea was that a judicious combination of different asset classes ensures a higher return per unit of risk. But this perception was seriously shaken during the financial crisis in 2008. At that time, the correlation between different investments increased sharply, prompting the somewhat cynical comment that diversification is something you have when you don’t need it (when volatility is low), but lose quickly when the pressure on the markets intensifies. The situation has not really improved since 2008, and the chart below suggests that even this free lunch is threatened with extinction. After all, since 2002 we can observe a steady rise in the average correlation among a wide range of asset classes, and recently the correlations have exceeded the spikes of 2008.

Source: BNP Paribas Investment Partners

There are several explanations for this development. One factor is the low interest rates compared with the returns sought by investors. In such a situation, investors will be tempted to respond to every return opportunity that presents itself, including short-term market movements, on the principle that every basis point of return helps. A second explanation is the shortening of the decision horizon, which means that even long-term investors take decisions for shorter periods, which they reverse, if necessary, at a later stage. This may suggest that they are less certain about the long-term forecasts and expectations, so they decide to play a shorter game. And thirdly, there is the growing popularity of instruments that enable very active investing. Exchange-trade funds (ETFs) spring to mind here. In the past, highly dynamic investment was only open to professionals via futures; now everyone has that opportunity via ETFs. 

These explanations actually provide more of an answer to the question of why volatility has increased, not why the correlation between asset classes has increased. One explanation for the stronger correlation is that these days, more than in the past, the shocks are global rather than country-specific. Shocks with a ‘local’ origin (e.g. the US housing crisis) will have a global impact through contagion. The same is true for the eurozone crisis. Here, too, the impact has gone far beyond the European Union, owing to a reduced risk appetite among investors, fears of a Europe-wide recession with negative implications for trade partners, and so on. 

The higher frequency of shocks provides a second explanation. Correlations increase when shocks occur. The above chart shows that there have been three shocks since 2008: 2008 itself (the financial crisis) and then 2010 and 2011 (the eurozone crisis in both cases), while the years between 2004-2008 were a relatively calm period. (Incidentally, the impact of shocks is accentuated on the chart because it uses a rolling average.) Once a shock has passed, the correlations weaken again, but when another shock occurs, they increase. More frequent shocks push up the average correlation. 

A third explanation involves an improved understanding of the behaviour of asset classes. We can call this the ‘generalised factor approach’. Equities provide exposure to an interest rate factor and a growth factor. High-yield corporate bonds do the same, which is why they show a close correlation with equities. Emerging market bonds have a domestic interest rate factor, an international (i.e. US) interest rate factor and a ‘global economic cycle’ factor (in short, optimism about growth generates more interest in these investments). The intensity with which these asset classes are exposed to the above factors will vary of course, but this does explain why even in bond portfolios, growth-sensitive bond categories perform better when confidence is growing. 

If investors are convinced they have a grip on these themes, then investing will become more and more a case of ‘surfing on beta waves’. That is to say, at one point the growth theme will dominate, and investments that rely heavily on this factor will increase in price; then another theme (inflation, monetary policy etc.) will come to dominate, and other investments will top the performance league. What matters for investors is to decide for each investment from their universe what factors are relevant and how strong a factor’s impact will be. Then they must decide which factor will be the focus of attention (‘what’s hot, what’s not’). The extent to which this factor is associated with economic growth will determine the markets’ risk-on or risk-off behaviour. Incidentally, momentum investing may enhance this phenomenon. In this situation, investors will buy an instrument because of its beta characteristics rather than any fundamental changes in the asset class. All this is then rationalised by arguing that risk appetite has increased or decreased... 

Going back to the chart, I would like to make two further comments. As the correlations surge, the chance increases that they will weaken again. In the past weeks the correlations have strengthened considerably, so we can assume that they will weaken again in due course. We can also see that the average correlations will weaken for a considerable time once the correlation spike has passed. Diversification will then re-emerge, but more than in the past, the behaviour of the various asset classes will be determined by several different factors. Investing will become more fascinating, because the single-factor approach will be less successful, but it will also be more difficult because many more data will need to be analysed. Still, that will be worth it, if only because diversification will be more effective. 

William De Vijlder
Chief Investment Officer, Strategy & Partners
BNP Paribas Investment Partners

Should inflation be feared as a result of unconventional monetary policies?

Text published in Beleggers Belangen on 9 March 2012

Will the ECB’s long-term refinancing operations (LTRO’s) create inflation? Yes, certainly… inflation in the number of articles on this subject. On 29 February, the day on which the European Central Bank launched its second LTRO operation (the first took place last December), Twitter was flooded by a tsunami of messages. They were mostly very critical and maintained that it “was encouraging speculation”, “creating a bubble”, etc. This negative viewpoint is astonishing. Where would we be today if the ECB had not implemented this courageous policy in December, thereby putting an end to the vicious circle going from public finances to increased financing costs, to the banking system, to economic activity, and back to public finances? Would the economic sentiment surveys be at the same level as at present? This is a rhetorical question. Nevertheless, a former director of the Dutch central bank sent a tweet posing the very relevant question of how to eventually drain more than one thousand billion euros of liquidity from the system. This will be a crucial question for financial markets in 2013 and 2014, and a similar debate is raging in the United States.

Regarding LTRO 1 and 2, it is important to distinguish between three phases. The first one, which is behind us, involved stopping the vicious circle described above. The second one aims to improve the operation of the mechanisms of monetary policy transmission by lowering the interest rates at which euro-zone governments obtain financing, reducing uncertainty and restoring confidence, all this with a view to a pickup in economic activity. This process takes time, but the financial markets anticipate it, which explains stock markets' exuberance in January and February. The rise in equity markets and in the prices of corporate bonds (and hence a fall in their financing costs) are therefore not the sign of a resurgence of speculation, but reflect a certain revival of risk appetite on the part of investors, which is a necessary and beneficial stage in this process of recovery. The third phase will aim at gradually draining the exceptional liquidity that has recently been injected into the economic system. This will be a necessary exercise, failing which there could be extraordinary growth in credit volumes followed by an acceleration of inflation.

It is important to qualify these two aspects. Regarding credit volumes, everything depends on the private sector's will to leverage further and on the banking sector's will to grant credit. These two behaviours are pro-cyclical: when activity picks up, credit demand and supply increase. Given the current excess reserves in the European and US banking systems, there is a fear that an increase in credit demand could lead to a sudden, sharp increase in the volume of credit in the economy, since reserves are there to respond to the increase in demand. In such a scenario, the central bank would risk losing control of the money supply.

Inflation, for its part, could appear following an acceleration in the volume of credit, but the risk, at least in the short term, seems limited. Unemployment in those countries/regions that have adopted unconventional monetary policies is and will remain high (just consider the long-term forecasts of the Federal Reserve, for example), so there will be no rapid resurgence of bottlenecks. However, monetary creation inevitably leads to more inflation: Without being a monetarist, Friedmanite theory teaches us that this type of operation drives up asset prices and the general price level. The theory is of course more complex, as is the reality. Monetary creation can, in particular, be absorbed by a fall in the velocity of circulation of money, or by a rise in the prices of certain assets (stock markets, real estate), thus limiting the risk of acceleration of "real" inflation, i.e. inflation in goods and services. Flight to the rest of the world is another important factor and can reduce the risk of domestic inflation. We note that analyses of quantitative easing policy are too often based on a closed economy. Now, we live in a globalised economy, where monetary creation in low-growth countries results in capital flows to emerging countries which have sounder economic fundamentals. These countries typically have a high level of capacity utilisation, close to or even above normal. Capital inflows push up the prices of financial assets there, which is equivalent to giving a boost to an economy that is already humming along. Unless currencies soar (cf. the 2010 debate over the “currency war”), we can conclude from this that monetary creation will more rapidly create inflation in emerging countries than in the countries with an unconventional monetary policy. This point, moreover, was raised recently by John Authers in the Financial Times, and I share his viewpoint.

Inflation in emerging countries will take the form of a rise in commodity prices, for example, or a rise in the cost of labour, and it is via this channel (commodities, import prices) that developed countries will experience an acceleration of inflation while at the same time having a high level of unemployment. Accordingly, investors should be prepared for a change of regime. In this new era appearing on the horizon, central banks will gradually prepare their exit strategy (this concept had already appeared, prematurely, in 2010), namely a return to a normal monetary policy to prevent a structural rise in inflation and hence the loss of their credibility. This transition poses several challenges for central banks (how to prevent the long end of the yield curve soaring as in the 1994 bond crash?), but also for investors. For the latter, the normalisation of monetary policy implies that attention will now be focused on a detailed analysis of companies' cash flows, an exercise that is far more difficult than managing expectations of monetary bazookas.

William De Vijlder
Chief Investment Officer, Strategy & Partners
BNP Paribas Investment Partners

Follow me on Twitter!

From today onwards, I will add comments on Twitter to complement my blog launched in 2008. If you wish to follow me on Twitter, click here @DeVijlder

Dossier : Asset Management - Trends in Asset Allocation

Dossier : Asset Management - Trends in Asset Allocation,  in tijdschrift (Bank en Financiewezen), jaargang (2008), nummer (6-7), website larcier (www.larcier.com), Uitgever : Larcier.

In recent years we have witnessed several new trends in asset allocation. The emphasis on liabilities has increased, the number of asset classes has risen significantly and there has been an increased willingness to invest in illiquid asset classes. The focus on the return distribution has increased as well and underpins the interest in hedge funds, alternative beta and investments with a convex return payoff. The growth of derivatives has boosted the trend towards the separation of alpha and beta: an asset class is now held because of its risk premium and an actively managed portfolio of idiosyncratic risk is held for the alpha or excess return it can procure in the hands of a skilled manager. The complexity of these developments has given rise to the growth of fiduciary management. In terms of tactical asset allocation, the interest in quantitative approaches to expected return modelling continues to rise, underpinned by an effort to reduce the influence of psychological factors and to increase the number of de-correlated positions (breadth) whereby the returns of pair trades are easier to model compared to using a qualitative approach. Despite all these developments, the basic challenges of the simple, old-fashioned asset allocation are as high as ever. They concern the need to estimate the expected return vector of the asset classes over short and long horizons, to calculate the correlation structure and to anticipate its evolution and to find a way of dealing with fluctuations in risk aversion and possibly illiquid markets.

Download Dossier_Asset_Management_Trends_in_Asset_Allocation (pdf)