In UBS’ view, 1994 is critical for guiding investing today. The key point about 1994 was not that US bond yields rose during a global recovery. But that the leverage and positioning built up in previous years, on the assumption that yields would remain low, then got stressed. The central issue, they note, is that a long period of lacklustre growth, low rates and easy money induces individual investors, funds, non-financial corporates and banks to reach for yield. In many cases, they gear up to do it. And as Hyman Minsky warned; in this way, stability breeds leverage, and leverage breeds instability.
Via UBS: 1994 Redux?
Sebastian Mallaby has written an excellent account of the 1994 bond market blowout in ‘Hurricane Greenspan’, chapter eight of his book ‘More money than God’ (Bloomsbury press, 2010). In his depiction of the legendary hedge fund trader Michael Steinhardt – he describes how hedge funds, and a range of other financial institutions, chased convergence trades from 1990-1993.
They played term carry (borrowing short term to buy long dated bonds within the US). They ran crossregional carry trades (borrowing in Germany or the US to buy Italian & Spanish bonds as these countries prepared for EU membership).
And they rushed to buy assets that were priced off convergence trades; emerging market property, peripheral banks. They even bought defensive growth stocks (with the idea that the PE on a defensive growth stock should converge to the inverse of the 10 year yield).
We argue below that the set-up today is very similar to that in early 1994.
The danger in these trades is that a cyclical recovery, especially a global cyclical recovery, will cause yields to rise and compel policy makers to withraw accommodation. And that this can induce an outsized reaction in all the convergence trades ultimately priced off treasuries, as leverage is removed.
This is why the central lesson from 1994 is that, after a long period of easy money, when a cyclical recovery kicks in and policymakers are setting to remove accommodation, at all costs avoid convergence trades and avoid assets that are priced off convergence trades.
And the popular convergence trades of the past months have been;
- Emerging market credit
- Emerging market property
- Southern European sovereign debt
- Peripheral European sovereign debt
- US mortgage backed securities
- US and global high yield debt
- Global defensive growth stocks.
So what brings us to think that we can use 1994 as a guide to investing for the rest of 2013?
In the section below we highlight several key developments from 1990-1995 and the comparison with the current situation;
The Fed ran a very easy monetary policy from 1990-early ’94 in an attempt to reflate the US economy in aftermath of the S&L crisis. We have seen lower rates & even easier monetary policy since 2009.
US growth remained lacklustre throughout 1990-1993, going through a series of moderate ‘mini-cycles’. We have seen even more lacklustre growth over the past four years.
US 10-year treasury yields fell from 9% to 5% from 1990-early 1994, as a recession and then disinflationary pressure pushed down inflation expectations. Treasury yields fell from 4.3% in 2007 to 1.4% in the summer of 2012.
US banks hoarded treasuries.
Lending remained lacklustre.
Corporates hoarded cash & paid back debt.
From 1990-1994 Capital flowed into emerging markets. Asia boomed. The former USSR saw large inflows also. Capital flowed heavily into emerging markets from 2009-11, although it then slowed in 2H11 & 2012 as the Fed ended QE2.
Credit spreads tightened from 90-94, and from 09-13.
Commodities remained in the doldrums from 1990-1994. This was unusual, given the strong capital flows into emerging markets. But the implosion of the military/industrial complex in Russia from 1989 saw domestic demand for commodities collapse. Russia then exported nickel, aluminium, palladium, platinum, copper and oil to get hold of hard currency. Commodity prices came under intense pressure. This contrast is with the 2009-13 period – where capital flows & restocking drove commodity prices higher from 2009-11, but where capital outflows, destocking and new supply drove prices lower in 2011/12.
Headline CPI trended down, persuading many that there was no cause for rate hikes. We have seen a similar trend from mid-2001.
The dollar trade weighted index range traded between 80 & 95 from 1990-1995. An interesting development was that the dollar weakened while the US economy recovered through 1994, and while the Fed raised rates 225bps. The DXY has been range trading in a similar manner, broadly between 75 and 90 since 2009.
The extended period of low rates and strong capital flows into emerging markets induced a huge build-up of leverage across financial & non-financial institutions on a global basis.
The strong flows of capital into emerging markets set off the procyclical growth dynamic we have described regularly.
Capital inflows induce central banks to print their own currency to buy the dollars coming in. Bank deposits rise, and banks lend to construction and engineering companies. Growth & inflation pick up. And with nominal rates sticky, real rates fall. That in turn incentivises procyclical gearing up to buy & build houses, inventory and general fixed capital formation.
The Asian tigers grew aggressively, and their stock markets boomed going into 1993. Emerging markets recovered in 2009/10, struggled into 2011/12 and then saw a patchy recovery until recently.
The problem with the reflationary process in emerging markets is that it sows the seeds for its own destruction. Because the low real rates in EM induce excessive gearing & fixed capital formation – compared to a more balanced allocation of capital, had real rates stayed steady above zero. This leaves misallocated capital, and the latent potential for bad loans to emerge when credit becomes scarce. It also causes a deterioration in the trade balance. Both make emerging markets increasingly dependent on capital flows to stay afloat.
In many cases, emerging market governments will react to rising inflation by attempting to restrict credit growth (rather than raising rates). The problem with this is that it incentivises US dollar borrowing.
Emerging market business finds it attractive to borrow in dollars when domestic inflation is rising, the domestic currency is appreciating, and domestic borrowing costs are higher than dollar funding. And it is even more attractive when the activity the loans are funding – from inventory building to FCF – sees price/cost rises.
But when the trends reverse – the domestic currency depreciates, the dollar funding becomes more dear, the inventory values fall – then emerging market corporates can find themselves squeezed. Very rapidly.
But it is not just EM. In the long history of financial crises, the ‘reach for yield’ during a slow growth and low yield environment has on multiple occasions set up the conditions for financial stress when yields eventually rose.
The book ‘More money than God’ by Sebastian Mallaby (Bloomsbury, 2010), gives an excellent description of the leverage and yield enhancing structures that built up in the 1990-1993 period, and the carnage inflicted upon that leverage in 1994. Some examples include:
- Bank & hedge fund carry trades – borrowing at the short end to purchase long dated bonds.
- Borrowing in USD and German marks to buy Italian and Greek long term debt
- Borrowing to buy high yield corporate debt.
- he use of interest rate swaps to generate yield enhancement.
- Leverage purchases of buy-to-let properties
We have also seen a significant build up in leverage over the past four years. Buy-to-let investment has risen strongly in the US/UK/Switzerland/Scandinavia. Retail investors have become heavily exposed to credit through mutual funds and credit ETFs.
Investors became very overweight long duration defensive growth and dividend yielding equities, at the expense of cyclical exposure.
Investors have left themselves highly exposed to any kind of cyclical rally outside the US, as well as within it. Valuations (as we noted here) are extremely varied.
As macro activity in the US accelerated, corporates stopped hoarding cash and started to seek to borrow to expand their businesses.
US banks, which had been hoarding treasuries, sold them to make way for increased corporate loans. Treasuries started to sell off.
The Fed then responded to the steepening curve and the improving macro conditions by raising rates by 25bps in February 1994. This came as a surprise to the market, which was not aware of the Fed’s internal deliberations. The transcript of the February meeting indicates that Fed members were wary of a 1988/89 style spike in inflation if they did not start the process of tightening.
Greenspan believed that the curve would flatten, as markets anticipated tighter policy moderated inflation expectations in the future.
But that’s not what happened.
The rise in rates instead dented the derivative trades predicated on no rise in yields, and it squeezed carry traders. That induced a more aggressive unwinding of treasury holdings, as leveraged carry trades unwound. And the Banks accelerated the sell off as they sold treasuries to make space for increased corporate lending. So the yield curve steepened over the year, with 10-year yields rising 306bps vs the 225bp rise in Fed funds.
An array of casualties ensued, from Orange county, California, that went bankrupt due to its exposure to a series of exotic interest rate swaps. To a number of prominent hedge funds – which saw extreme losses in February 1994.
Then there was the international fallout. The sharp increase in domestic demand for credit, combined with the increase in real rates induced powerful capital flows back to the US. This sucked liquidity out of several emerging markets, whose central banks had to retire domestic currency to repay the dollars exiting their countries. Soon, countries that had seen the most aggressive investment booms, which had done the most aggressive US dollar borrowing, and which suffered the largest current account deficits, came under intense duress.The Mexican peso crisis erupted, and the seeds were sown for a sustained deterioration in Asia, before the full collapse of the Asian crisis in 1997.
One of the conundrums of 1994 was the US dollar. It would be logical to think that, with a sharp rise in US growth, in rates & yields that the US dollar would have rallied. But it didn’t. It fell.
An important reason was that the US recovery, while stronger than expected, was not a big surprise. But what was a surprise was the European recovery – after the sustained post-unification funk in Germany, and the Scandinavian banking crisis in 1992. In our view in commodity strategy – it was the relative surprises – which made Europe’s recovery much more unexpected, that triggered the currency move.
This is particularly interesting today – with the broad consensus that the US dollar is going to rally, due to the more robust recovery in the US and the potential for tapering.
But it is always worth keeping an eye on relative macro surprises.
We see the potential for a counter trend fall in the US dollar.
Now there are clearly some stark differences between today and 1994. Back then interest rates were much higher. So 300bps on treasuries increased rates by three fifths. The same rise from the July low would treble rates. And certainly, the authorities are first talking about an extended period of QE tapering. We are still a distance away from actual rate hikes.
The Fed is also much more transparent than it was under Greenspan in the early 1990s.
Where conditions are similar is that a very large structure of leverage has built up on the back of low rates, from leveraged property & credit buying, large retail exposure to yield enhancement products (high yield ETFs etc), earlier dollar leverage driven investment booms in emerging markets.
So where are we now. It looks to us very similar to February 1994.
The Fed’s continued insistence on talking tapering despite the recent rollover in US macro surprises has started to unsettle leveraged yield enhanced positioning.
The US high yield ETF has come under severe pressure. The US mortgage spread has blown out relative to the US 30-year treasury yield. South African and Indian currencies are under pressure. India has responded by raising taxes on gold imports.
In 1994, Mexico was the first to feel the brunt. Followed by South Korea in 1997. In 2013, South Africa is feeling the pressure. Although other emerging markets, notably China, continue to benefit.
The next big question is; can the US withstand a higher cost of capital, like it did from 1994-98.
In short, no!
In the mid-late ‘90s, the US coped with a higher cost of capital in several ways. It enhanced productivity through a rapid adoption of tech. Corporates geared up, which ensured strong liquidity growth and ‘efficient’ balance sheets. Corporates went through a second round of ‘just in time’ inventory management and outsourcing. Consumers benefited from the strong dollar and falling commodity prices – seeing their disposable incomes improve. And the disinflation in EM translated to a downtrend in yields from 1994, which allowed for an acceleration in the housing market and an expansion of household debt.
But we have a number of concerns that hint at vulnerability.
The first is that the potential for sustained disinflation over multiple years is less, because yields are already low. Consequently, there is less scope for a sustained recovery in housing – beyond the initial flurry of demand from rising household formation. The sharp rise in mortgage spreads is one hint that this transition may be more difficult. The spread on mortgages may be particularly important for the leveraged buy-to-let investors, who have been heavily involved in the recent surge in housing sales.
Because we understand that a large part of the buying is from investors then seeking to rent out the properties, we suspect that the follow-through consumer demand may not be as aggressive as previously imagined. If a household buys a house, taking on debt, it opens the floodgates to increasing debt fuelled buying of cars, household furnishings and white goods. A very different psychology comes from paying a month up-front on a rental. You are much more likely to cut back, to be more frugal.
A more subtle point is that the extended expansion of government spending as a share of GDP in response to the financial crisis is crowding out the private sector, and reducing the productive potential of the US economy. This stands in stark contrast with the tight control of government debt in the early 1990s under the Clinton administration.
These suggest that it is much less likely that we see the US enter a ‘high plateau’ of growth as we saw from 1995-98, where the US saw a powerful productivity & credit fuelled boom while the emerging markets deflated. And it makes it more likely that the US stays on a lower trajectory, interspersed with periodic recessionary slowdowns in the years ahead.
The point at which the market realises this would likely herald a significant risk-off event.