he twin barrels of financial innovation and globalization have significantly complicated the forecasting of asset returns in recent years. Two domestic bubbles in the last decade are testimony to the power of levered money and the recirculation of price insensitive reserves back into U.S. financial markets. While the late 1990s equity bubble had its roots in bonafide technological innovation, peak prices were driven by credit creation fed by globalized repatriation of Asian reserves following its financial crisis in 1997/1998. Similarly, what now appears to be confirmed as a housing bubble, was substantially inflated by nearly $1 trillion of annual reserve flowing back into U.S. Treasury and mortgage markets at subsidized yields, as well as innovative funny money mortgage creation which allowed anyone to buy a house at escalating and insupportable prices. Bond, stock, and real estate trends then, have recently been increasingly at the mercy of relatively price insensitive and levered financial flows as opposed to historical models of value or the growth of the real economy itself. An exorbitant P/E of 25x meant nothing in 1999 as unforeseen flows and investor mania drove them to 35x at the peak in 2001. An unchanged 10-year yield range centered around 4½% mystified the Fed, but not cash rich foreign central bank buyers as short term rates were aggressively raised by 425 basis points from 2003 to mid-2006. This foreign repatriation produced artificially low yields, (perhaps 50-100 basis points confirmed in numerous Federal Reserve staff reports and speeches in recent years) which in turn drove housing values to unsustainable levels as recently as six months ago – the estimated peak in national home prices.
Investors these days, therefore, must turn their analysis inside out or perhaps calculate in reverse – counting not from 1 to 100 but 100 to 1, as if there were beer bottles on the wall and they were being taken down one-by-one and passed around until drunk. But whether beer or $5 million dollar houses in Newport Beach , my critical point is that asset prices are no longer entirely a function of the real economy: it can be just the reverse. The real economy is being driven by asset prices, which in turn are influenced by financial flows of non-historic origin, composition, and uncertain longevity. What used to be an Economics 101 “CIG + exports-imports” analysis leading to predictions for interest rates and stock prices has turned into an Economics 2007 analysis of corporate buybacks, international reserve flows and hedge fund/private equity positioning seeking to front run or take advantage of the first two. And it’s not simply a question of analyzing the animal spirits or “exuberance” of investors wherever they may be. Corporations are buying back stock with their historically high profits not really because they’re enthusiastic about their own company’s value, but because they have little else to do with the money. Likewise, foreign central banks and petroreserve recyclers are turned on more by capping their own currencies or geopolitical considerations in the Middle East.
Investors have no more significant example of the influence of financial flows on asset prices than tracking the pace of the U.S. trade deficit in the 21st century, as good a point as any to mark the beginning of our new financial era, since it encompasses both equity and housing asset bubble peaks. In effect, despite the chicken and egg aspect of why the trade deficit exists – because foreign investors want to invest in the U.S. or because U.S. consumers want to buy things – there is likely near unanimity that it is now responsible for pumping nearly $800 billion of cash flow into our bond and equity markets annually. Without it, both bond and stock prices would be much lower, the $800 billion for instance representing 3 - 4x our current federal budget deficit. Almost perversely, then, an increasing current account deficit supports and elevates U.S. asset prices as the liquidity from it is used to buy stocks and bonds. Chart 1 shows the trend of this dynamic pump since the beginning of the century.
Notice that in 2001 a monthly trend reversal of $10 billion ($120 billion annually) neatly coincided with a 20% decline in stock prices and a flat bond market despite a developing recession. While the real economy was certainly influencing stocks and corporate profits, the effect of financial flows was also becoming apparent, certainly in the bond market and suspiciously so in stocks as well. The draining of $120 billion from the foreign cash flow pump appeared to have magnifying, “it’s different this time,” effects on both. The trade deficit resumed its downward trek in 2002 and as it did, stocks recovered and a strange phenomena began to be observed in the bond market – what Ben Bernanke was to call a global savings glut – which because it was recycled primarily into U.S. financial markets via an accelerating trade deficit – resulted in artificially high bond prices and low interest rates. Bernanke’s “glut” effect appears to have continued up until the present day as unusual and persistent negative yield curves in the U.S. have had little of their normal cyclical dampening effect on asset prices and the real economy.
Although the above historical analysis is subjective and vulnerable to “sampling error” (economist speak for too short a modeling timeframe) there is an inherent logic to it: more money in the “bank” – asset prices go up; fewer deposits – asset prices go down or perhaps up less. The logic no doubt is complicated, however, by the continued thrust of financial innovation and the willingness on the part of investors to take additional risk with even the same pot of money. Financial derivatives of all (almost unimaginable) varieties have sprouted to the surface in recent years allowing homeowners to lever home prices, institutions to compress risk spreads, and almost all assets to occupy a seemingly permanently higher plateau based on increased liquidity and perceived diversification of risk across the system. I have my doubts about this permanent plateau, but the market seemingly does not, and there’s no doubt therefore that “flows” are not the entire answer, even if they have increased their level of influence in recent years. If risk takers decide to “play,” there is a casino awaiting them 24-hours a day, trade deficit or no trade deficit.
With that important caveat, let me proceed to analyze another source of increased cash flow that has markedly influenced asset prices in recent years. I refer to corporate profits and their meteoric rise since the 2001 recession, increasing from 5¼% to nearly 9% of GDP as shown in Chart 2. While normally much of that rise of over $400 billion after tax dollars (almost identical to the pump provided by our increasing trade deficit over the same period) would have been reinvested in physical plant and equipment, this time it was not.
Because of more attractive yet limited alternatives in Asia much of the “windfall” was used for share buybacks. Chart 3 displays buyback activity on a quarterly basis since 2003, revealing an average increase of (surprise) nearly $300 billion annually that has gone to support financial asset prices via the stock market.
Combined, the total rise in corporate share buybacks and the financing for bond and stock markets via the increasing trade deficit have injected an average of perhaps $1 trillion annually of purchasing power into our asset markets since the end of the 2001 recession. Because hedge funds and levered players of all types have been aware of this trade deficit/share buyback “put” and have acted upon it, the incalculable but conservatively estimatable pump from these two sources alone have poured in several trillions of purchasing power per year. Take that money and use it to invest in further high powered and levered financial instruments such as CDOs, CPDOs, and 0% down funny money mortgages of all varieties and you can understand why asset markets have done so well in recent years, and why, as my initial Outlook sentence suggested, it is so hard to analyze “value” in asset markets these days. Prices are increasingly being determined by value insensitive flows and speculative leverage as opposed to fundamentals.
But there’s an ill wind blowin’ this time around, or to put it another way perhaps, many of our proverbial 100 bottles of beer on the wall may have been taken down, drained, and have totally inebriated the asset markets to the point of preventing further significant price advances. Drunks do, after all, at some point stagger home, roll into bed, and at least sleep it off for a good number of hours. The suggestion of no more bottles of beer on the wall comes from several sources, the first of which appears in Chart 1 as a recent reversal in the trade deficit. While some of this improvement is due to the standard dollar weakness of the past 12 months and its dampening impact on imports, much of it is due to the decline of oil since August/September of 2006. Follow with me if you will a projection by PIMCO analyst Ramin Toloui in Chart 4 that depicts the change in trade flows at a given dollar price of oil.
As you can see, the recent $20 reversal in per barrel oil prices results in a reduction of $100 billion or so in the annual trade deficit, and a like amount of liquidity extraction from bond and stock markets, much more if associated leverage is unwound. Granted, some would claim that there will still be $700 billion or so of purchasing power coming into our markets, but higher asset prices in a levered economy are dependent on greater and greater injections of liquidity, not less. Should oil hold in the $55 range, this extraction of high powered 100+ proof alcohol from the markets will be noticeable.
The second source of vulnerability comes from the corporate buyback stash, a trend itself as Chart 3 points out that is beginning to level off and reverse. Peter Bernstein, in a recent January missive, suggests that corporate profits as a % of GDP cannot continue to grow at the same pace. “Everybody else” he writes “is going to want a piece of that juicy action. Employees will demand higher wages, customers will demand lower prices, and the government will levy higher taxes.” Bernstein’s forecast undoubtedly has a several years timeframe, but there is a more immediate influence in the air shown in this Outlook’s final Chart 5.
Corporate profits are significantly influenced by the growth rate of real and (importantly) nominal GDP. As Chart 5 hints, should nominal GDP decline into the 4-5% range over the next several years as discussed in last month’s Investment Outlook, corporate profits and ultimately the juice for share buybacks will be affected as well. Chart 5, points to the possibility of reducing profits as a % of GDP by as much as 2% over the next several years if 5% nominal is where we are headed. If so, then share buybacks could be cut back by a good $100+ billion in the near term future.
I have gone on too long and maybe drunk a few too many of those intellectual beers myself I suppose. But if I’ve made a case for cash flows of non-historic origin and their associated leverage affecting asset prices to the upside in recent years, it’s fair to analyze what might happen when the brew begins to run dry. How many beers are left on the wall? It’s hard to know and close to impossible to pinpoint especially so because of our financially oriented casino offering innovation after innovation. That is the nature of today’s financial marketplace. But as Alan Greenspan stated in August of 2005: “The determination of global economic activity in recent years has been influenced importantly by capital gains on various types of assets and the liabilities that finance them.” If financial innovation and the resultant leverage derived there from are reflective of our 21st century liability creation, and if the U.S. trade deficit and corporate share buybacks are responsible in part for asset capital gains, then asset prices are increasingly being determined by flows and the leveraging of those flows, as is economic growth itself. There may be some bottles yet left on the wall, but I would be careful. The risk markets (including bond term premiums) if not drunk, are definitely not walking a very straight line. Stocks, credit spreads, and yes intermediate and long term bonds relative to a likely unwavering Fed Funds rate in 2007’s first half, may stagger shortly.
William H. Gross