April European Issue Focus




Global growth remains slow…

World GDP growth remains sluggish. We forecast growth of just 2.3 per cent in 2016, down from 2.5 per cent in 2015 and well below the long-term average of the last thirty years of 2.8 per cent.

Meanwhile, the volatile performance of global financial markets in Q1 2016 has sparked concerns that the world could tip into recession. We examine this risk using a range of key indicators and by comparing the current situation to historical episodes (see Table 1).

…and some indicators worrying

G7 industrial output: On the real economy side, the data reveal that weakness in world output is mostly concentrated in industry. We estimate that industry in the G7 countries (U.S., UK, Japan, Canada, Germany, France and Italy) as a whole will enter a technical recession (two straight quarters of decline) in Q1 2016. For the U.S., Q1 data already showed a second successive quarterly decline in industrial output.

However, industrial recessions in the G7 are more common than world recessions (which we define here as periods where per capita world GDP shrinks for two successive quarters). We identify eleven G7 industrial recessions since 1973, but only six world economic recessions, implying that the probability of a G7 industrial recession being associated with a world recession is 55.0 per cent. Notable cases where G7 industry has sent a ‘false signal’ include mid-1998 and mid-2012.

Equity prices: Turning to financial market indicators, the most well-known is equity market slumps, which can risk recession through a variety of channels, including household wealth effects and effects on consumer and business confidence. World equity markets slid sharply in the first six weeks of the year before recovering, and remain well below their peaks of last May (see Chart 1).

A common saying is that the equity market ‘predicted nine out of the last five recessions’. Our analysis broadly supports this view. Since 1973, world stock market slumps of 6.0 per cent or more over six months have only been associated with recessions half the time. Recessions generally have large equity slumps, but not all large equity slumps are associated with recessions.

Once again, there have been notable periods when stock markets have sent false signals, including 1987, 1998, 2002 and 2011-12.

While it is correct that equity slumps can send false signals, we cannot dismiss the risk from falling stocks. In 1987 and 1998, the global monetary authorities reacted to equity price slides with strong monetary stimulus, which arguably helped limit the economic fallout. In 2011-12, recession was again avoided but world growth did slow notably to only 2.0 per cent year over year in Q4 2012 and Q1 2013 from around 3.0 per cent a year earlier. Moreover, the U.S. Federal Reserve launched QE3 in late 2012, helping spark a renewed equity rally.

High yield corporate spreads: The next indicator we consider is high yield credit spreads, which have widened sharply over the last six months to their widest level since 2011 (see Chart 2).

Insofar as these are an indicator of credit conditions being faced by firms, they may prove to be a useful forward-looking indicator of world output. They may be an indicator of ‘financial accelerator’ effects, whereby falling asset prices worsen corporate balance sheets and reduce access to credit, feeding back into weaker investment and output.

If we look at how U.S. high-yield spreads behave as an indicator of world recessions, the record is relatively good, although not perfect.

Since the mid-1980s, we have identified five periods where junk spreads spiked by 300 basis points or more to levels above their long-term average (excluding the current episode). Of these, three were associated with world recessions, making for a 60.0 per cent ‘hit rate’.

The two false signals were in 1998 and 2011-12, where sharp increases in spreads (to around 500bp and 800bp, respectively) were quickly followed by big rallies. Spreads have also fallen back from their recent peaks in the current episode, but nevertheless remain elevated. Moreover, in the 2011-12 episode, although there was no recession, world growth did slow sharply. So spreads at current levels are a risk to growth.

Bank credit standards: Another important set of indicators, in our view, are surveys of bank credit standards. A number of central banks run these, but the one with the longest history and which encompasses several periods of global recession is the U.S. Senior Loan Officer Survey.

These surveys, especially the survey run in the U.S., have a relatively good record in predicting movements in world output. Since 1990, we find eight occasions when the survey has spiked into ‘tightening territory’. Of these, five have been associated with recessionary periods, a ‘hit rate’ of 63.0 per cent.

The latest reading of the U.S. survey, for Q1 2016, showed a net 8.0 per cent of U.S. banks tightening credit standards for corporate borrowers. This was the second straight quarter of net tightening.

However, the current reported tightening balance is quite modest by the standards of recession periods; the average for such periods is around 40.0 per cent. There have also again been periods where this indicator has got it wrong. Spikes into tightening territory in 1996, 1998 (when the spike was large, if short-lived) and 2012 were not associated with recessions (see Chart 3). Finally, similar surveys run by other central banks are showing a more benign picture, especially in the Eurozone.

Commodity prices: Our final indicator is commodity prices. Commodities may react quickly to shifts in global demand, with a sharp slump in their prices signalling flagging world growth. Real non-fuel commodities are currently around 5.0 per cent lower than a year ago and 40.0 per cent below their 2011 peak.

Our analysis suggests that commodities are the least reliable of our indicators. We find eleven instances of real non-fuel commodity prices dropping 10.0 per cent or more over six months since 1973, but only four of these cases were associated with world recessions.

The problem here is likely to be that supply factors can have a big effect as well as demand factors; in our view, the recent weakness of global commodity prices has reflected both supply and demand factors.

‘Yellow alert’, not red alert

If we consider all our indicators together, we would argue that while some have shown worrying developments over recent months, none are yet flashing red. Moreover, we would note that these indicators are imperfect, having sent ‘false signals’ in the past. The recent rally in financial markets also cuts the recession risk (see Table 1).

Overall, we think the current situation most closely resembles those in 1998 and in 2011-12. In neither case was the worsening of key indicators followed by world recession. However, world growth did slow sharply in 2012 and 2013, with the G7 countries coming close to recession in H2 2012. So, we would argue that the risk of world growth undershooting our expectations and slipping below 2.0 per cent this year is high.

It is also possible that financial market conditions could worsen again, driving up recession risks once more. There are a number of possible catalysts for this; corporate profits could disappoint, driving down stocks; corporate defaults might rise quickly, leading to widening corporate bond spreads; and there are various risks from the political side, such as in Russia, the Middle East and China.

This update was researched and written by Oxford Economic Forecasting Ltd., 121 St. Aldates, Oxford, OX1 1HB, England, as of 29 April 2016.