Originally posted on February 19, 2017
I have a feeling that equity markets are setting a trap for investors, but I can't quite figure out which kind it is. Will the last bull be sucked in before the disappointment sets in, or are we now on a sustainable glide path towards new highs with maximum frustration for the sceptics? We didn't get any decisive clues last week. Equity volatility rose a tad, but ranges remain incredibly tight across a number of key asset classes. False breaks are guaranteed, and vol-sellers will continue to play cat and mouse with the heroes trying to straddle the ranges, playing for a breakout.
I have told two stories on these pages in the past few months; firstly, I have argued that equity investors would be best served by fading the reflation theme, and rebalance towards bonds and defensive sectors. Secondly, I have raised the alarm on risk assets as a whole, not in an attempt to catch the "big sell-off," but based on a number of my models and indicators. Holding bonds and defensive equities have started to pay off, but benchmark equity indices have continued to push higher. On the face of it, the macro data support this move. Survey data in the U.S. have been spectacular, and this week's batch of EZ data likely will confirm that that solid momentum in Q4 carried over into Q1.
Is global growth accelerating?
Headline leading indicators certainly suggest that something is brewing. The U.S. NBER LEI rose 2.5% year-over-year in January, the strongest pace since December 2015, and leading indicators in the Eurozone, Japan and China are also edging higher. In addition, I have thrown in the latest update of Goldman Sachs' global leading indicator for good measure. It also is telling an upbeat story. The message based on these indicators is clear; global GDP growth is accelerating, although momentum in the LEIs is still modest compared to previous cyclical upturns.
To get a more detailed picture, I decided to dust off an old model that I have used in the past; a diffusion index of macroeconomic leading indicators . The first chart below shows my core index of normalised leading indicators, which is simply the average of the individual economies' z-scores. Not surprisingly, it is broadly consistent with the global PMI. But the upshot is that it is smoother, which we can exploit to derive a momentum indicator. I have done that in the second chart.
It is difficult to deny the key message here. Global leading indicators began a broad-based upturn in the middle of last year-before Brexit and Trump-and it showed no signs of abating at the end of last year. If the early 2017 data are anything to go by, it will be a similar story in Q1. In addition, the momentum of my leading index suggests that the global PMI will remain at about its current level at 53 in the first half of the year. Admittedly, this prediction is based on a rather arbitrary "fit" between the two indicators, but as long as the momentum index is rising on a six-month basis, the outlook should be decent. A similar message comes from looking at a standard diffusion index, which counts the number of indices rising on a year-over-year basis, plotted here against global GDP growth. The index-which takes a maximum value of 25-jumped to about 16 in Q1, indicating the odds are indeed for accelerating global GDP growth in the short run.
Even if global growth is improving the key question for equity investors is whether the good news is already priced in. Macroeconomic surprise indices have stormed higher, indicating that markets have run ahead of the improving economy. But what does the leading indicators discussed above have to say? To answer that question I ran a little study. Firstly, I regressed six-month returns of the MSCI World on the three-month momentum of my leading index. Secondly, I calculated the residual to look for extremes. The intuition is that of a simple error-correction model. Predictions of returns are likely to be most inaccurate at extremes, and we can exploit this to show periods of large deviations between actual trailing returns and predicted returns based on macroeconomic data. Another way to describe the residual is that it captures all the non-macroeconomic factors which determine the six-month change of the MSCI World. The chart below shows the residual and the trailing returns of the MSCI World.
The conclusion from the charts above is rather more favourable to the bulls compared to the idea that surging macroeconomic surprises will lead to disappointments ahead. In fact, in January the model based on macroeconomic news-using December data pushed forward one month-suggest that trailing returns of global equities were a tad too low. In any case, it is far from previous extremes, which would have allowed investors to make a solid bet on mean-reversion in equity returns based on either good or bad macro news flows. The analysis above supports the idea that equities could be destined for better times, at least based on an improving macroeconomy. But, as I have showed, the trailing performance of cyclicals relative to defensives definitely suggests that chasing outperformance in materials, industrials and financials won't work in the short run.
Allow to me throw in two more pieces of evidence to increase the confusion. Firstly, we need to think about where we are in the business cycle. Developed markets and their economy have come a long way since the crisis, especially if we focus on the U.S., where the key equity benchmarks tend to drive markets in the other major OECD countries. The first chart below shows the level of unemployment claims to ram this point home, but I could show a number of indicators to suggest that the business cycle is in its later stages. I am not so silly as to suggest that claims will go up just because they're low. The business cycle does not work that way, and growth undoubtedly remains decent. But inflation is creeping higher and the Fed is anxious to move, which provides another important hint. The key here is to watch the U.S. yield curve. I suspect it will flatten, not steepen, in response to the Fed nudging rates higher, adding to my conviction that the business cycle is getting long in the tooth.
Secondly, we have emerging markets. I think the key story here is the asynchronous business cycles between EMs and the OECD, and that currently provides opportunities. But even here, a hiccup could be in store. The second chart shows that real narrow money growth in China has collapsed. This is partly because of a slowdown in nominal M1 growth, but it is also because of higher inflation. I more than willing to accept the idea that China will "muddle through," but if it does it will also be accompanied by a "stop-go" business cycle driven by the PBoC's willingness to "allow" the economy to expand, via its control of the credit flow.
Consensus trade du jour: overweight EZ equities
I usually confine myself to very close inspection of the EZ economy in my day job, but last week I also commented on the growing choir of experts on the tee vee extolling the virtues of buying Eurozone equities. At a first glance, this doesn't look like such a stupid idea. Indeed, I suspect the main thrust of the argument is based on something akin to the following chart, which I think comes from 13D research.
I can't really refute the contrarian attractiveness of this setup, but I remind everyone that relative and absolute performance is not the same thing. That said, I have sympathy for the idea of buying EZ equities because political risks are overestimated and because financials-which have been a key source of EZ underperformance-are facing better conditions. Markets always have been horrible at pricing political risks in Europe. They get it wrong every single time. My main problem, however, is that EZ equities aren't really cheap in an absolute sense and the idea that earnings automatically will catch up with high multiples, in turn justifying a further rise in multiples, and so on, isn't really convincing to me. We know that y/y comps for energy and materials will be very favourable in Q1 and Q2. But this has been obvious for ages, and isn't really a good argument to pile in. If we look at German industrials and materials, for example, which arguably would be at the core of a long-EZ strategy based on strong macroeconomic performance, they are not cheap. In fact they look quite expensive to me.
It is the same if I look at "quality" mid-cap firms in the Germany and France with strong margins, good free cash flow and solid revenue growth. Most of them are trading on very high trailing multiples. As such, I am unconvinced that EZ equities offer the opportunities the consensus thinks they do, not because I think the EU/EZ will fall apart, but because they look rather expensive.
Rear-ended by Syntel…again
The portfolio had finally started to stretch its legs a little bit, even nudging ahead of the MSCI World, but then came the black hole of Syntel earnings. The firm beat the street's estimates, but guidance was poor and punters, who had been pushing up the price going into the report, jumped ship. It slid a cool 18% on the day of the release, and the sell-off continued the following session with a relatively "modest" 3% drop. This is the second quarterly earnings season that the portfolio is being taken to the woodshed by Syntel. So what gives?
I am not sure. But I calm myself with the glaring divergence between trailing returns of Syntel and my valuation score.
More interestingly, though, I am in increasingly exquisite company. John Hempton from Bronte Capital used his latest blog post to ask questions on Syntel. Full disclosure: I am lucky to be in regular contact with John, and I have had the pleasure of drinking my sanity away on obscure bars in Madrid, while he was lecturing me and my friends on the virtues of a Canadian aircraft landing gear manufacturer. When this guy digs deep, he means it. Not only will he visit the factory floor, he will also make it his business to learn how every single widget works.
This time around, though, he is asking for help. He started with the assumption that Syntel is a "fraud", and thus a short. But he has since changed his mind finding no obvious evidence of foul play. I am not sure I have much to offer here actually. Syntel is trading is a deep discount to its own history and peers, and has strong fundamentals. That is just about as long as I will go. My assumption after Q3 earnings was that the market was discounting the need to raise equity as a result of burning capital. After, if we are late in the cycle, paying a "special dividend" is not a good idea. Indeed, the firm's capital management strategy seems odd in the extreme. For the record, here are some bearish and bullish views from the comments John has garnered so far.
"Very few and concentrated customers suggests lumpy and somewhat unpredictable growth. Lack of a significant sales effort means very low sales expenses and higher margins. Frugal company run almost as a family run entity. Nothing all that exciting to analyze but just a buncha penny pinching cheapskates so all good employees leave as soon as they can - hope this connects the dots for you."
"Regarding the margins, my conclusion is that Syntel operates with a culture of thrift. They don't pay their executives aggressively, and they spend less on marketing. The smaller sales budget is likely what is causing their subpar revenue growth. I think they got away with being thrifty on SG&A for several years because their top customers (AMEX, State Street, and FedEx) were growing spending at a solid clip. So Syntel was able to approximate industry average revenue growth without the marketing budget. That has come to an end."
It will be interesting to see how this story evolves. Elsewhere, I have reluctantly covered my Home Depot short. I think the thing is way too overlevered for its own good, but the cost of borrowing was becoming too expensive and the thing recently broke out of range to the upside. It had to go. Otherwise, I am trying to do as little as possible. Not buying vol in a sideways market usually is the clever move. Meanwhile, the stare-down between different fractions in the market continues. It's bulls vs bears and vol-sellers vs vol-buyers. I imagine a lot will be decided when either of the sides blink.
 - I know you don't like black boxes, so here is the recipe. The core index is based on monthly macroeconomic leading indicators, or in some cases "GDP proxies," from the major economies. My index has input from 25 economies. I have tried to use national indices where possible, but I have had to default to OECD LEIs-amplitude adjusted indices-in many cases. Most national indices are rebased as a GDP-like index, which requires you to use a growth rate, while the OECD indices are already de-trended around 100. I have used z-scores to normalise everything to make them comparable. Once you have done this, there are many ways to "diffuse" the information. But six-month momentum, a count of rising/falling, z-scores etc are useful transformations.