All Weather Strategy: Bridgewater Associates

Winner: Best global macro hedge fund

In a world of inconsistency, Bridgewater’s All Weather Strategy stands out. Although the usual investment warnings apply, the strategy has certainly bucked performance trends.

In 2011 it returned a stunning 19.0% and over the last three years 19.9%. Slightly less stunning, but still acceptable in today’s climate. The strategy managed a positive annualised 8.6% return over the last five years and 9.8% over the last 10 years.

[Pictured: Leo Lueb, portfolio construction strategist, Bridgewater Associates]

From inception through June 2012, it produced an annualised return of 9.4% with 12.4% volatility and a Sharpe ratio of 0.52. Not bad, particularly since the expected return for the strategy is 7.8% above cash.

Since it was first conceived, the strategy has not fundamentally changed. Currently with $56 billion under management, it is “quite a bit away” from any capacity problems, as long as the portfolio follows its rebalancing rules, says Jim Haskel, a senior portfolio strategist at Bridgewater and a senior member of the research group. He notes that the size of the markets traded helps keep transaction costs low.

The reason for the strategy’s success is straightforward: balance.

All Weather is Bridgewater’s optimal strategic asset allocation strategy. Instead of generating returns through trading active views, the strategy seeks to collect asset class risk premiums in an environmentally balanced way.

Its diversification is based on Bridgewater’s understanding of the structural relationship between asset classes and different economic environments. Bridgewater has identified assets that naturally diversify each other based on their fundamental relationship to changing growth and inflation environments. The strategy allocates risk exposure with the idea of creating a portfolio that is balanced to perform well regardless of what happens in the world economy.

Asset classes include nominal bonds, inflation-linked bonds, sovereign and corporate credit spreads, equity and commodity markets. A broad variety of instruments are used to implement the strategy. These include, among others, exchange traded futures contracts, over-the-counter derivatives, cash securities and spot and forward contracts in the international currency market.

“What we basically determined through many years of research at the firm, going all the way back to 1975, is that the most important driver of cashflows of any particular asset is growth and inflation, with growth being the volume of economic activity and inflation the pricing of that economic activity,” explains Haskel.

“Each of those drivers has an influence on each asset’s cashflows. So if you take an asset like Treasuries, every day they are discounted for the future rate of growth and future inflation rate. What happens every day is that either growth rises or falls below what is discounted by the markets or inflation rises or falls above or below what is discounted in the markets,” he continues.

Those influences are roughly equal in terms of how they drive that asset over time. So for Treasuries the strategy would put them in a falling growth environment. If growth falls below what was previously discounted because a weak employment report is released, for example, all things being equal, that will be good for Treasuries.

The same idea works on the inflation side. “Sometimes the effect on Treasuries, for example, can be muted because you’ll have a falling growth environment but a rising inflation environment. Sometimes the magnitude of the contribution of each one can be different and, therefore, either Treasuries will go up or down a little bit but [inflation and growth will] have contradictory influences,” he adds.

Balance
Bridgewater draws what it calls ‘environmental boxes’, allocating assets it believes are reliable in a series of environments: rising growth, falling growth, rising inflation and falling inflation. By doing that, All Weather achieves a balanced asset allocation that produces good absolute returns independent, as much as possible, of the underlying economic environment and how that environment shifts.

“All of the assets that we use in our All Weather Strategy are assets that we‘ve been dealing with for a very long time, even before we started the strategy,” says Haskel.

“The beauty of it is that those assets have basic characteristics that do not change very much over time. Any tweaks we make to the strategy are really around things like how to construct the best benchmark of equities in terms of how we hold equities or how we hold inflation index-linked bonds or how we hold nominal bonds around the world. The mix of them or where we implement on the yield curve – that kind of thing,” he continues.

Haskel will not go into the instrument-level allocations of the strategy. However, what he does say about the strategy is that it is difficult to attempt to get a balanced asset mix if there is no fundamental understanding of the drivers of those assets. The only way to do it without this understanding, he contends, is to use some quantitative method.

All Weather uses this fundamental understanding to achieve balance. “If you think about environmental drivers of assets, you never have to predict correlations. The correlations will always make sense because they are simply a reflection of whatever the underlying drivers were,” he says.

As an example he looks back to the 1970s when the most dominant influence in the broader economy was inflation. “In the 1970s inflation was discounted to be very low and by 1980 it was discounted to be very high. There was a change in discounted conditions,” he notes.

The nature of the cashflows of stocks and bonds are such that they do not respond well to a big change in discounted conditions driven by inflation. Therefore, they both did poorly over the decade and were positively correlated.

“Now if you look at the last decade. What do we know? In 2000 at the height of the telecom and technology bubble in the stock market, if you actually worked your way back to what that implied for earnings and what kind of growth rates would have had to be true to produce those earnings, basically growth would be discounted to be very high,” he explains.

“Ultimately what happened over the next 10-12 years – where we are today – is that growth has actually come in much, much lower than was previously discounted.”

This he believes influences nominal bonds, such as Treasuries, in a very positive way. But it also influences things such as stocks in a very negative way because profits are cut.

So those two assets have been inversely correlated over the last decade. “If you can understand how the assets respond to changing inflation and growth conditions, then you don’t need to predict correlations because correlations simply explain what happens in discounted conditions,” Haskel concludes.

Despite the recent market turmoil, the strategy “performed as expected in terms of staying balanced to the changing economic conditions. Its returns were a bit better than expected simply because central banks have made it a good time to own risky assets by virtue of keeping cash rates extremely low into the future.”

Tweaking
Changing the composition of the portfolio, Haskel admits, is rare. He restates the concept of ‘tweaking’ the portfolio. The starting point is that there should be an equal amount of risk from the four sub-portfolios.

“From there we look at each sub-portfolio and we ask ourselves the following question: which asset class is or can be most reliably consistent if we get that type of environment,” explains Haskel.

For example, because current bond yields are low from a historical perspective and when this happens there can be lower volatility on the upside of bond prices than on the downside, the strategy has shifted some of the implementation points to the longer end of the yield curves in order to get balance. “That’s just an implementation change. That’s not a change in what we’re holding in the strategy.”

The reasoning behind this construction comes from stress-testing the logic over time and over many countries while factoring in liquidity dynamics. This is done for every sub-portfolio. For example in rising growth, equity makes up the biggest composition of that sub-portfolio. But the sub-portfolio also includes corporate spreads, emerging market debt spreads and commodities, but with less weight than equities.

For falling growth the most reliable assets to hold would be nominal bonds and inflation-linked bonds because they both have a real yield component. In such an environment real yields compress, prices go up on bonds and yields fall.

In a rising inflation sub-portfolio, the strategy holds things such as inflation index bonds and commodities, the first because they pay out in inflation and the latter because they are the source of inflation as well as emerging market debt spread. In a falling inflation environment, nominal bonds and equities are held. Each asset is held in a fixed percentage that is rarely changed.

“You can see that most assets are held for both growth and inflation balancing. We assign a risk weight to them. We combine the rising and falling growth and inflation assets and that’s our overall weighting for assets in the All Weather portfolio in terms of risk,” explains Haskel.

“Then we work out from there what kind of capital allocation we need depending on the implementation choices in order to achieve that risk. That’s how it’s put together,” he says.

“If we make a change it’s more of a tweak. I say ‘tweak’. Because it’s minor relative to the most important aspect, which is being balanced. It may be that instead of implementing bonds at the 10-year part of the curve, we implement it at the 15-year part and therefore have to own less in order to get an equivalent amount of risk. Those kinds of changes. But those are literally almost not even worth commenting on.”

Bridgewater’s world view

Research is a large component of Bridgewater’s edge. Jim Haskel is not only a senior portfolio strategist, he is also a senior member of the research group with expertise in portfolio structuring and foreign exchange, interest rates, commodity and equity markets.

The objective of the research department is to understand how the world works, according to Haskel.

“If we understand how the world works, how the global financial system works, if we literally understand that machine, then we can make sense of the conditions in the world and those conditions will produce outcomes that will show itself in asset class pricing,” explains Haskel.

That understanding is applied to the only two strategies run by Bridgewater: Pure Alpha and the All Weather. As it pertains to All Weather, the understanding simply reaffirms how the strategy works and how the characteristics of asset classes correspond with the underlying environment. “All we have to do from there is that, when drift occurs in the portfolio, we just rebalance to our target weights.

“That’s why we believe the return stream that is holding risky assets is really a great return stream. You don’t have to predict where the world is going. That’s what‘s exciting to us.”

Nevertheless, Bridgewater believes that the most dominant force driving the world economy is the deleveraging occurring in the over-indebted western world. This is a world where borrowers are now stuck with debt, interest rates are near or at zero and the ability of central banks to lower the debt servicing costs on debtors is no longer as easy as it was when you could just lower interest rates.

This, says Haskel, happens every 50-75 years in capitalist societies. When it happens the business cycle is no longer the dominant influence. However, the long-term debt cycle is turning from an upward path and now points to deleveraging.

“Spending is just about income or borrowing. If borrowing is being reduced because debtors are trying to reduce their debt, that puts the western world in a very precarious position in which we have very low growth rates and are vulnerable to recession or even depressions while this debt is being worked off,” says Haskel.

Printing money
Traditionally the way to get out of deleveraging somewhat successfully is to have the right combination of austerity, transfers of wealth, restructuring of debt and printing money. When applied to areas like the periphery of the eurozone, there is no easy way to do this in lieu of lowering interest rates.

To restructure debt, countries need to go through some measure of austerity and an equal proportion of printing money. The problem, says Haskel, is that Europe has too much austerity and not enough money being printed or money transfers. That is creating what Haskel terms a “dangerous dynamic” in which the market fears the eurozone will eventually come to an end.

In the US and the UK there is a better mixture, simply because they have independent monetary policy and can continue to stimulate the economy by printing money. This has been done in the US through two rounds of quantitative easing and Operation Twist.

This has produced a broader mix of austerity transfers from the federal government to states and restructuring of debts, particularly in the mortgage area. As result, even though the outlook is painful, it is not nearly as painful as the depressions that exist in parts of Europe.

“If the deleveragings are handled well, defined as a roughly equal mix across those four areas, we’re looking at a multi-year process,” concludes Haskel. “Historically these things have basically lasted anything from 10 to 20 years and I see no reason why this time should be different.”

However, if there is a disproportionate amount of austerity, for example at the expense of other things, then that could push countries into a great depression and a vicious cycle which could last equally as long and be even more painful.

Conversely, if there is a disproportionate amount of printing of money, it could be a shorter timeframe but the results could be an inflationary spiral, which extinguishes the debt overhang more quickly, but at great pain and often leads to social dislocation.

Bridgewater’s central expectation is a multi-year process that could last a long time and, according to Haskel, “we’re only in the first three to four years of this” period.

Looking at the eurozone crisis specifically, Haskel believes it has so far been a “very reactive process” partly because it is a monetary union made up of 17 countries without a single sovereign identity. What is needed in order to get effective transfer mechanisms is some sort of fiscal or federal union similar to the US where the centre can transfer funds to the states as needed.

The European Union is starting to develop such mechanisms but it is politically an unpopular move. Nevertheless, the implication of Europe coming undone is “so bad is that what we’ve seen whenever it comes up to the brink, the Europeans act to buy more time. It’s been very messy and so far very inadequate but also enough to get the job done for now. But much more will have to be required, particularly if Spain and Italy, which are two very big countries in the eurozone, are to be able to extract themselves from their debt situations in a successful way,” says Haskel.

Many more transfers from the richer countries in Europe – particularly Germany – than have been agreed so far will be needed as well as more easing by the European Central Bank in order to place these countries on a sustainable debt path.

Risky environment
In such an environment investors holding risky assets should not contemplate any sort of asset allocation strategy that produces absolute return other than the model used by All Weather. This is because many strategies are holding risky assets and are overly concentrated in a single asset. For Haskel that is the last place an investor should be in an environment where the divergence around a central outcome is much wider than normal. “I want that balance more than ever today. No question,” he states.

“As it pertains to Pure Alpha, I think even though in this strategy we’re trying to predict where things are going, we’re doing it in a very diversified way. We are not putting too much risk on any single thing. Essentially we are trying to generate a winning percentage north of 50% because we know that when we do that we will perform roughly at expectation,” he continues.

“That’s our goal: it’s to stay balanced across the world in various assets, trying to extract alpha in little pieces and get more wins than losses so that no matter what happens in Europe, or for that matter globally, we should be able to produce absolute returns.”

The expectation at the moment is that the developed world will continue to go through a difficult period as US growth continues to slow. With such a low growth expectation, the economy is vulnerable to any sort of exogenous shocks.

The good news is that there are some signs, because of the nature of the deleveraging, that things such as housing are starting to recover. If housing were to get going a little bit, that would help credit creation and also help drive some growth.

“In the US we’ve got a very weak level of growth. In the absence of any exogenous shocks we think we could see stable if not faintly higher [growth] as we move into 2013,” he says.

But Haskel cautions that the US is “bumbling along on the bottom” with the periphery of Europe in a depression and Germany slowing. European companies that export outside the region are doing relatively well, which should result in some follow-on growth. Nevertheless, European growth is expected to continue to be weak and vulnerable to a shock as the deleveraging continues.

Interestingly, Haskel says China is a “concern”. While it does not have a sovereign debt overhang, it has leveraged up a lot over the last few years because its export partners have been weak. The government has produced growth by creating a lot of internal debt.

“What we are worried about in China as we measure what’s going on in the economy is a pretty hard cyclical softening of conditions. That is worrisome if not handled correctly. That could create a problem,” Haskel says.