Available evidence suggests that the preliminary first-quarter GDP estimate released on 25 April will show a small rise from the fourth quarter, although there is downside risk stemming from March’s unusually cold weather.

The Office for National Statistics (ONS) has released data up to February for industrial and construction output and to January for the dominant services sector. If seasonally-adjusted output in each of the sectors were to remain at its latest-available level, GDP would be unchanged on the quarter, with tiny output rises in services and industry offset by a fall in construction.

There are, however, grounds for believing that services output will build on a 0.4% January gain in February / March. First, average retail sales volume in February / March was 1.7% above the January level. The sales numbers are used to measure retail trade output within the index of services. Retail trade has a weight of 6.9% in the index, implying a 0.12% boost to services output in February / March.

Secondly, turnover figures for February released today suggest a further rise in output of other private services industries. The turnover survey is a key input to the index of services but the published numbers – unhelpfully – are in current prices and unadjusted for seasonal and calendar effects. The chart compares services output with a seasonally-adjusted version of the turnover series – the relationship is imperfect but the monthly change in turnover has directionally matched that of output in nine of the last 10 months, and was positive again in February.

A reasonable assumption, based on the above, is that services output rises by 0.4% in February and is unchanged in March. Absent any revision to earlier data, this would imply a 0.3% quarterly gain, in turn suggesting GDP growth of 0.2% (assuming, as before, that industrial and construction output is unchanged between February and March).

This “forecast” is vulnerable to weakness in services industries not covered by the turnover survey (e.g. government and finance) and could also be upset if the ONS has early evidence of a depressing impact from March’s cold weather, although any such effect will unwind in the second quarter.

GDP growth of 0.2% would be little cause for celebration but the recent pattern of revisions suggests that the preliminary number will be raised over time – quarterly GDP changes since the start of 2009 have increased by an average of 0.15 percentage points (i.e. comparing the latest data vintage with preliminary estimates).

The MSCI World equity total return index in US dollars is within touching distance of its 2007 all-time high, having risen by 131%* from a post-crisis low reached in March 2009. This strong performance, despite a disappointing economic recovery in the Group of Seven (G7) major countries, is attributable to three main factors.

First, equities were attractively valued following the 2007-09 bear market. According to a forecasting model** employed by US economist and fund manager John Hussman, for example, US stocks were priced to deliver a 10-year return of 9.2% per annum (pa) at the March 2009 trough – see first chart. The prospective return, admittedly, reached much higher levels in the 1970s and 1980s – it peaked at 20.0% in June 1982. The March 2009 forecast, however, was the strongest since 1992. Secondly, global growth has been respectable since 2009, despite a sluggish G7 recovery, reflecting the rapid expansion and increased weight of emerging economies. The IMF’s world GDP measure, calculated using “purchasing power parity” country weights, rose by 4.1% pa over 2010-12, above an average of 3.4% since 1980. The IMF expects continued respectable growth of 3.3% in 2013. World stock market earnings depend on GDP performance globally rather than in the G7.

Thirdly, the liquidity backdrop for equity and other markets has been unusually favourable since 2009, partly reflecting extreme and unconventional monetary policies. As previously discussed in these commentaries, a key liquidity indicator is the gap between the annual growth rates of G7 real (i.e. inflation-adjusted) money supply and industrial production. Faster expansion of real money than output implies that there is “excess” liquidity available to flow into markets and push up prices. This condition has been met in 34 out of 48 months since March 2009.

What do these considerations suggest about current equity market prospects? The trend rise in global GDP and corporate earnings should continue to be supported by emerging-world catch-up. The real money / industrial output growth gap, meanwhile, is still positive, though has narrowed in early 2013 as monetary expansion has moderated and the global economy has regained speed after a slowdown in mid-2012.

Equity market valuation, however, is now much less attractive than in 2009. According to the Hussman model, for example, US stocks are priced to achieve 10-year performance of only 3.0% pa, well below the historical average – first chart. Mr Hussman, therefore, has been suggesting that investors should hold cash rather than equities until an opportunity arises to lock into a less unfavourable long-term return.

Such an argument may be valid for US stocks but carries less force in other markets, where valuation appears reasonable. Applying a similar model to the MSCI World ex US index, for example, suggests a 10-year prospective return of 8.3%.

A further consideration, of course, is the lack of appeal of other investment options – particularly government bonds and cash. The prospective 10-year return of 3.0% pa on US stocks is 1.3 percentage points above the current yield on US 10-year Treasuries*.

Many fixed-income investors, moreover, achieve exposure via  funds that maintain a constant proportion of long-term securities rather than buying a 10-year bond to hold to maturity. They are, in other words, exposed to a risk of capital loss if either real yields revert to their historical average or inflationary expectations rise. A Hussman-type forecasting model suggests that a bond fund always investing in the current 10-year Treasury benchmark will deliver a negative return of 0.2% pa over the next 10 years, assuming that the real yield*** rises from its current -0.6% to 2.8% and inflationary expectations remain unchanged – second chart.
US equities, in other words, may be priced to deliver a modest return but still appear attractive to bonds. The two models suggest an equity return premium of 3.2% pa, above an average since 1960 of 2.6% – third chart. Valuation is the key driver of long-run performance but shorter-term equity market movements usually reflect the global economic cycle. A strong rally since mid-2012, for example, anticipated a pick-up in economic momentum in late 2012 and early 2013. This pick-up, in turn, was foreshadowed by faster global real money expansion from spring 2012 – the real money supply leads economic activity by about six months, according to the monetarist rule.

As noted earlier, real money growth has moderated recently, having peaked in late 2012. Allowing for the six-month lead, this suggests that the economy will lose momentum from mid-2013, a development that could trigger another bout of weakness in equity markets. This forecast is provisional  because monetary trends are not yet signalling a major slowdown and could revive, partly reflecting ongoing policy stimulus from the Federal Reserve and Bank of Japan.

Summing up, equities are much less attractive than in early 2009 but are nevertheless likely to outperform cash and government bonds by a significant margin over the long term. Investors seeking to deploy new cash, however, may wish to delay pending a possible economic slowdown later in 2013, which may create a better entry opportunity.

*As of 17 April.
**The model assumes growth in dividends, earnings and nominal GDP of 6.3% pa – based on post-war experience – and mean reversion of the ratio of market cap to GDP over 10 years.
***The real yield is calculated by subtracting the consensus 10-year inflation forecast from the current 10-year Treasury yield.

Chinese industrial output is expanding respectably, nothwithstanding a negative market reaction to March data released earlier this week.

Annual growth in output fell to 8.9% last month, below market forecasts and the lowest since August 2012. The annual number, however, continues to be depressed by weakness in spring / summer 2012. Output is estimated* here to have risen by a seasonally-adjusted 5.1% in the six months to March, or 10.4% annualised.

Both new order flows and monetary trends suggest that growth will be sustained at around its recent pace into the second half – see chart and Friday’s post respectively.

*The estimate also adjusts for the timing of the Chinese New Year.

Global industrial output is conservatively estimated* to have risen by 2.6% in the six months to March, or 5.2% annualised. This would be the fastest six-month growth rate since March 2011 and compares with a small contraction in September 2012 – see chart.

Recent stronger output validates the “monetarist” forecast here that the global economy would regain speed in early 2013, notwithstanding ongoing fiscal tightening, in lagged response to faster real money supply expansion between spring and autumn 2012.

Global six-month real narrow money expansion has slowed since October 2012, with the decline estimated** to have continued in March. This suggests that industrial output growth will moderate in mid 2013 but real money trends remain respectable by historical standards, with a temporary fall in inflation providing support, courtesy of lower commodity prices.

The gap between real money and output expansion is a summary measure of “excess” liquidity available to boost asset prices. The gap remains positive but narrowed further in March, consistent with the liquidity backdrop becoming less favourable at the margin – a development possibly explaining recent choppier markets.

*”Global” = G7 plus emerging E7. The March estimate uses data for the US, China and Russia and an official survey forecast for Japan, with output elsewhere assumed to be unchanged from February.
**The estimate uses data for the US, Japan, Brazil, China and India and assumes unchanged six-month growth elsewhere.