Wadhwani was speaking at the conference, because he manages an onshore UCITS version of his strategy for GAM, GAM Star Keynes Quantitative Strategies. Of particular interest to the audience was how GAM and Wadhwani went about this, as the UCITS III directive precludes the use of commodities, a key part of the Wadhwani offshore fund’s portfolio.
Investment approach
Wadhwani started his presentation by saying that he has always been very influenced by Keynes’ view of how markets work, and that has had an important impact on his models. Essentially, he said, Keynes emphasised three key points.
Firstly, Keynes said that we know remarkably little, and as we try and form an assessment of the future, we often confront something more like pure uncertainty rather than just risk. Second, given that, it is only natural that there is a lot of ‘herding behaviour’, people try and hug the consensus. This explains why momentum strategies can work for a while. Momentum represents an important part of Wadhwani’s quantitative approach, but not all of it. Third, Keynes warned about the perils of momentum strategies when he talked about ‘animal spirits’ and how these could lead to sudden lurches in the market.
Therefore, if one is going to use momentum as a strategy, it is very important to combine it with other things which help the fund manager or investor to go to the outer edges of the herd and then peel away a significant distance from the herd before the stampede starts. For that reason, Wadhwani’s modelling approach, like a lot of well known trend following funds, uses momentum and other closely related technical indicators. But he combines it with a variety of different types of non-price information. In particular, he uses economic fundamentals, where he is principally concerned with the growth and inflation outlook relative to the consensus view, sentiment, flow of funds, and explicitly allows for inter-market linkages. “Of course we worry about valuation too,” Wadhwani said.
To illustrate why this approach can help, recall that in the first half of 2008 the market made a low in March. Most trend followers would have been short global equities at that point, which happened to coincide with the Bear Stearns rescue. However, Wadhwani’s models, despite having an important momentum component, were actually neutral at that point. The profits that had been made by being short global equities in the first part of 2008 were not given back when the bear market bounce occurred.
To make this point more generally, Fig.1 shows the Japanese equity market from 1992. “We actually created this picture in response to a request from a client who wanted to know how we would fare in a world which alternated between being optimistic and pessimistic, but perhaps around a downward trend,” Wadhwani told the conference. “Now that of course is where Japan has been over its two lost decades.”
The bottom line is a passive strategy of being long only back in 1992. Said Wadhwani: “At the time I was global equities strategist at Goldman Sachs and certainly much of my peer group back then was recommending buying the Japanese equity market, because it had fallen so much, and on some valuation indicators looked good. Of course, if you did that, it ended in tears.”
Alternatively, an investor could have had the foresight to go short, and that might have clearly enhanced his wealth, but it has done so with a considerable degree of volatility, because there have been at least six bear market rallies north of 30% over that period.
The line above that is a momentum strategy, and Wadhwani admits it gives him some comfort that a simple momentum strategy outperforms either passive strategy of being long or short. It has of course been characterised by a great deal of volatility. The final line is what Wadhwani is setting out to try to do: this overlays momentum with information about sentiment and the business cycle. “That certainly would have helped you navigate Japan rather better than momentum because of this alternating optimism and pessimism,” he explained.
Adapting the approach for UCITS III
When Wadhwani set out to build a UCITS III fund for GAM, his firm was not just looking at the existing regulations: “We were saying to ourselves that, in a ‘bash the banker’ type of environment, where the authorities are likely to become more and more restrictive on a three to five year view, we needed to come up with a product that was likely to be more immune to future regulatory change. In some respects, we were quite conservative.”
At the moment the fund offers weekly liquidity and aspires to offer daily liquidity. Fortunately everything Wadhwani invests in within the offshore fund is highly liquid, so in that sense his strategy was a natural fit for the UCITS strategy. The main restriction that he faced related to the rules about commodities. From his perspective, the big choice he needed to make was to trade commodities using swaps or ETCs, or to drop them from the portfolio?
Another key consideration for Wadhwani was that he had no desire to launch a UCITS product if it was going to be considered in the marketplace in any way inferior to the existing offshore product. They could be different, but they had to be comparable in terms of quality.
In terms of using swaps, which was the route he would have to take if using replication, the disadvantages were higher costs, counterparty risks, and for him the key point, exposure to future regulatory risk. “I’m not saying this based on a one year view, but on a three to five year view the hostility towards the whole financial sector is so considerable that one should not assume that you aren’t going to get more and more regulatory interference,” Wadhwani explained.
Dropping commodities from the portfolio would mean some loss of diversification benefit. But Wadhwani is running a systematic macro fund, and the macro factors that help drive commodity prices also drive the other asset classes. Secondly, if the fund is going to continue to invest in equity markets and not commodity markets, it can continue to get exposure to the commodity bet via equities that are closely related to commodity markets. Third, Wadhwani was much influenced by the fact that theoretically, a priori, some asset classes are more amenable to modelling using systematic macro factors than others. “We know that the half life of reversion back to fair value is a much shorter period for fixed income than it is for commodities,” he told the conference. “That heartened me, because it pointed to a way forward, whereby a UCITS fund could focus on fewer asset classes where our models had a higher average alpha.”
This leads to an offshore fund, which trades a broader set of asset classes, but where on average the models have a lower average alpha. In addition, in Wadhwani’s offshore fund, he insists on diversity; he does not necessarily have the optimal portfolio weights. “We reward diversity: some asset classes get a higher weight than would be implied by the optimiser,” he said.
The UCITS fund by contrast has fewer models which both theoretically and empirically have a higher average alpha. There are therefore two different products. The UCITS fund is distinct from the offshore version. “Hand on heart, I don’t know which one is going to do better over the next five years – they are just different,” Wadhwani said. “Indeed, if you look back, they have pretty similar reward/risk ratios.”
Wadhwani has been running the offshore fund for a while, and the issue has been one of deriving a track record for the UCITS because of these restrictions. He had to back out of the returns what the UCITS track record would be, and then ask Ernst & Young to review it.
In Table 1 the first column is the illustrative track record of the GAM UCITS fund, the second column is the actual track record of the offshore fund. The information ratio is very similar, while the Sortino ratio is a bit higher for the UCITS funds, but actually the return-to-drawdown ratio is higher for the offshore fund. They can be seen to be quite highly correlated at 0.77. Said Wadhwani: “I can genuinely say to clients that you have got two different products, quite highly correlated, but ex ante I can’t tell you which one is going to do better over the next five years. Ex post, they have been pretty similar too.”
Columns 3-5 show the performance of three well-known CTA indices. If Wadhwani’s approach of placing a great deal of emphasis on non-price indicators has any empirical validity, it should actually show up in superior return-to-drawdown ratios and superior Sortino ratios. The Sortino or the return-to-drawdown is generally better for both the Wadhwani entities than it is for the indices. “Comparing ourselves to an index is quite a stiff test as indices already have significant diversification benefits,” Wadhwani said. “If you combine a medium term trend follower with a short term fund, which some of the indices do, you’re already naturally boosting your Sortino ratio. Even against that stiffer target we seem to hold up pretty well.”
Why the medium-term outlook may help
Conventional assets don’t do well either during inflation or deflation. They need relative macro certainty and price stability, and this is why Wadhwani thinks alternative assets are going to be much more important than conventional assets on a five to 10 year view; the most likely scenario is a very significant degree of macro instability.
“There is a lot of uncertainty now about medium term growth and inflation,” he said. “Long only strategies are likely to perform poorly; you are therefore going to need an agile investment approach. I would suggest that momentum is going to outperform a lot of conventional passive strategies. More general approaches, which include non-price, should outperform trend following, and these are now beginning to be made available with a UCITS wrapper.”
Dr Sushil Wadhwani, CBE, is the Founder and CEO of Wadhwani Asset Management LP, a London-based asset management company that specialises in systematic macro investing. He is responsible for managing GAM Star Keynes Quantitative Strategies, a systematic macro fund within a UCITS III framework. Wadhwani was a full-time member of the Monetary Policy Committee at the Bank of England in 1999-2002. Prior to this his roles included director of research, head of systems trading and a partner at Tudor Group, and director of equity strategy at Goldman Sachs. He began his career as an academic in the Working of Financial Markets department of the London School of Economics.
