In recent blogs in this series, I have described the immediate outlook for US GDP growth as “surprisingly strong”. By coincidence, Jim O’Neill, my ex-colleague at Goldman Sachs, wrote a piece for the FT on Tuesday in which he argued that the strength of the recovery in the US economy would be one of the surprises of the year. These assessments have been seen by some as far too optimistic. Clearly, the US economy remains plagued by excessive debt and a chronically under-employed labour market. Furthermore, in a longer term context, the present recovery has not been sufficient to reverse the slow growth rate in the US economy in the past decade. So I have been re-assessing the case for “optimism” on the US.
Many observers of the financial system believe that market economists prefer to make optimistic forecasts, because this approach “sells better” to clients than pessimism. To an extent, this may be true, though I believe sufficiently in human nature to conclude that this tendency is mostly subconscious, rather than deliberate. Actually, my own observation is that economists and traders who specialise in fixed income, often have a bearish bias on the US economy, driven by their worries about the high levels of private and public sector debt in the economy.
In equities, though, optimism reigns. After all, since equities tend to rise over long periods, while they fall more sharply over shorter periods, optimistic analysts will be right more often than they will be wrong (even though, when they are wrong, they will be very wrong). The same tendency, incidentally, also afflicts many funds managers, who find that long positions are far easier to hold, in career terms, than short positions. All of us in the financial markets need to watch out for these biases in our own behaviour.
Whatever their underlying biases, there can be little doubt that financial market economists have a much shorter term outlook than academic economists or policy makers. This is because the markets themselves react so sharply to very minor fluctuations in economic variables. For example, since last summer, the probability of a double dip recession in the US during 2011 has perhaps dropped from about 25 per cent to about 10 per cent. The probability-weighted difference between these two outcomes for the equity market should be fairly negiligible, since the long term path for profits and discount rates would not be very different between the two.
Yet equity markets have risen by over 20 per cent in the past 6 months – a rally which would be very painful for any asset manager to have missed. Clearly markets were significantly “surprised” by this turn of events, which is why market economists spend so much effort in trying to understand minor fluctuations in the economic cycle. Like it or not, they bring with them very large opportunities for profits or losses.
Still, in the end, these small cyclical fluctuations will be of no consequence compared to the more enduring trends in underlying economic growth. John Ross, currently visiting Antai College in Shanghai has argued in his excellent blog that there is no excuse for optimism of any kind about the performance of the US economy, because the underlying growth rate has been falling for several decades, and even a fairly strong cyclical recovery in the imminent future would not dent this depressing trend.
The first graph 
shows 20-year and 10-year trailing growth rates for US GDP since the late 1960s. As Professor Ross points out, there has been a persistent tendency for trend growth in the US economy to slow down progressively since the 1960s, and in fact he argues that this has occurred decade by decade for over a hundred years. This is true, though I think that some of this is due to a slowdown in population growth rates.
The second graph shows the 10-year trailing population growth rate in the US, along with the per capita growth rate in GDP, which is what really matters for the average living standards of the nation.
It is clear that population growth slowed from about 2 per cent in the early 1960s to about 1 per cent in the early 1970s, and that it has remained roughly at this rate ever since. This explains a good part of the slowdown in GDP growth. In fact, once we adjust for population growth, the per capita GDP growth rates did not change very much from the mid 1970s to the mid 2000s. Over that period, they seemed to be stuck fairly close to the 2 per cent per annum rate which growth economists like Robert Barro have argued is normal for advanced economies.
However, even on this basis, there has been a large drop in the 10-year growth rate lately. Something has clearly gone very wrong. The key question now is whether we are simply seeing the effects of a temporary (though very severe) recession, or whether something more profound and more permanent has happened. In the past, the US economy has been fairly good at “shaking off” periods of recession, so real GDP per capita has usually returned to its pre-recession growth path over a period of years. But I readily concede that it may be more difficult this time, given the high levels of private and public debt, and a very low net savings ratio by historic standards. And I also agree that no-one could conceivably describe the current state of the labour market as satisfactory.
The battle for a sustained recovery will be fought over many years, not just the few quarters which normally dominate market sentiment. Genuine optimism about the US probably needs to wait until quite a few more skirmishes are won.
