Recent research from the World Gold Council is particularly pertinent following the revision to the US First Quarter GDP figures.
Posted: Wednesday , 04 Jun 2008
The most recent figures from the United States suggest that the first quarter GDP came in at +0.9%, an upward revision from the flash estimate of 0.6%. At first pass this is encouraging, but closer inspection shows a rather gloomier picture. Much of the growth in the real GDP figure was due, in the words of the US Department of Commerce, to an upturn in inventory investment that was partly offset by a deceleration in the Personal Consumption Expenditure (PCE)". Positive contributions also came from exports of goods and services, federal government spending and private inventory investment, while the durable goods PCE was negative, as was residential fixed investment (which should not come as a surprise), which dropped by 26%. Inventory adjustments contributed 0.2% to the GDP figure while the automotive sector shaved off 0.4%
Without the underpinning from durable goods, therefore, and with the latest Consumer Sentiment Index registering a 28-year low, the words "stagflation" and "recession" are again populating the vocabulary of market commentators. This makes one of the latest pieces of research from the World Gold Council partially interesting.
Natalie Dempster, the Investment Research Manager at the Council, has produced a paper that studies the performance of gold in recessionary conditions and one of the primary conclusions is that regression and correlation analysis suggest that there is no relationship between changes in US GDP growth and changes in the gold price. A US recession, therefore, would not have negative implications for the gold price. This is a reflection of the unique drivers of the gold market, and underpins the advection of gold's role as a diversifying asset, even in times of recession.
The World Gold Council analysis notes that both macro data and the Federal reserve's rapid loosening in monetary policy, both in terms of interest rate shifts and the massive injections of liquidity, suggest that the US is "at serious risk of, or possibly already in, a recession". The technical definition of a recession is two consecutive quarters of negative growth in real Gross Domestic Product, although the National Bureau of Economic Research (NBER) Business Cycle Dating Committee definition is of "a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales" and in truth the debate over the most accurate definition could fill many pages. Perhaps we should be thinking in terms of "perception" of a recession, since perception is so important in defining activity - and this is why the Consumer sentiment Index is currently so important. The latest survey from Reuters / University of Michigan came in at 51.1, the lowest since July 1979 when it was 44.1. Over the following twelve months, real consumer spending dropped by 0.3%, and the figures suggest that we are heading for a downturn again this time around, although probably somewhat shallower than in 1979/80 (when gold, incidentally, was hitting the then record of $850/ounce, although there were a number of other forces at work other than just economic - including an oil shock).
The Council research points out that there are obvious winners and losers in terms of asset performers in a recession. While car manufacturers and homebuilders, along with financial stocks, suffer as result of reduced consumption and slowing bank lending, defensive stocks such as biotechs or foodstuffs tend to perform well, as do fixed-income assets as interest rates are lowered in an effort to stimulate consumer demand. Commodities tend to underperform on the back of reduced demand.
There have been five recessionary periods since gold was released from its price peg in 1971, on the basis of the NBER definition. Two have each lasted 16 months, namely November 1973-March 1975 and July 1981 - November 1982; two have been of eight months' duration each, from July 1990 - March 1991 and March 2001 - November 2001; the fifth was just six months from January to July 1980. There has been no clear pattern in gold's price performance. Between November 1973 and March 1975 it gained 88% and in the second 16 month recession it rose by 7%. In the first eight month recession the price declined, as it did in the six month recession in early 1980 while in the recession following the bursting of the dot.com bubble it rallied by 7%.
The study notes that there is "no relationship between the gold price and economic activity full stop", with slowing growth sometimes accompanied by a falling gold price and on other occasions, most notably in recent months, gold has rallied strongly during a slowdown.
Regression and correlation analysis back up this statement. The correlation coefficient between quarterly GDP growth and the gold price from 1973 to 2007 is a paltry 0.02 (on a scale of zero to one). Similarly a ten-year rolling correlation coefficient between the two (where the scale is measured from minus one to plus one) has fluctuated merely between -0.2 and +0.2 - with the exception of two quarters in 1983. This, given the number of observations in the analysis, is effectively zero.
Compare this with other assets; the study points out, for example, that the correlation coefficient between ten-year bond futures has been "continuously negative and statistically significant" for much of 1972 - 2007.
Backing this analysis with an overview of the characteristics of gold's supply and demand dynamics, this is a useful study to illustrate how gold does, or more to the point does not, react to changes in GDP and, thus bends to the Council's argument for the use of gold as a diversifying asset within a portfolio.