Are Markit PMIs Really Reliable?

Includes: EWG, FXB, FXF, URR
by: George Dorgan

In a previous post on Seeking Alpha, we compared Purchasing Manager Indices (PMIs) with stock market valuations. Our conclusion was that especially German stocks (NYSEARCA:EWG) are overvalued, because PMIs were indicating a contraction of the German Gross Domestic Product (GDP). The GDP together with interest rates and inflation is the main determinant for foreign exchange markets.

When Forex traders saw the eurozone GDP forecasts on Monday morning, the euro performed very strongly against all other currencies, but stocks lagged. After the release of a commentary of the business survey provider Markit on the Eurostat's GDP data, we decided to examine the PMIs of this company in more detail. Markit showed in their commentary the accentuating divergence of Northern and Southern European countries which might continue for some years.

Looking into details, another divergence, namely the one between the Composite Purchasing Manager Index (Composite PMI) and the GDP for Germany and France became apparent.

Purchasing Manager Indices

The Composite PMI (Purchasing Managers' Index) "is produced by Markit and is based on original survey data collected from a representative panel of manufacturing and services firms." PMIs strongly influence foreign exchange rates and stock markets. Most recently they were one trigger of the strong fall of the euro against nearly all other currencies, especially between April and end of July.

Markit has a quite good track record in predicting GDP. However this time they seemed to get something wrong. If the Composite PMI were true, then German GDP would have contracted by 0.5%. In reality GDP rose by 0.3%, a miss of 0.8%. For the French Composite PMI the divergence is not that strong, but Markit missed by 0.4%.

Are Purchasing Manager Indices really leading indicators ?

What we all learned is the following:

Purchasing Manager Indices are leading indicators, whereas GDP is a lagging one.

Looking at the graph of the German GDP above I really cannot see the empirical evidence that PMIs are leading indicators. In Q2 2008 even the opposite occurred: the German GDP contracted but the composite PMI was unable to detect the upcoming recession, the PMI pointed to continuing expansion.

PMIs do not reflect government spending and inventories

In Q2/2012 German GDP rose thanks to the strong current account surplus in state and private consumption, whereas investments slowed. According to the Markit Chief Economist Chris Williamson French GDP was buoyed by government and inventories.

PMIs are sentiment indicators

One should never forget that PMIs are based on surveys and are subject to human sentiments, even if these purchasing manager sentiments are more reliable than consumer sentiments (see more).

Everything is relative to past experience: German manufacturing purchasing managers compare the current situation with the strong expansion in 2010 and 2011, when huge Chinese and European demand asked for more and more German machinery and cars. This is finished for now. Consequently the German manufacturing PMI (value: 43) currently drags down the composite PMI to 47.5 despite a German services sector in expansion. But the weak value of 43 seems to be overblown similarly as in 2011 when the German PMIs predicted a lot stronger GDP growth (see GDP vs. PMI graph above).

In the UK or the United States we see a different picture: Both Markit services PMIs are expanding, in the US even the Markit manufacturing PMIs. According to the Markit Global Composite PMI the "UK reported a reduction in output for the first time since April 2009" in the month of July. This implies again a big miss, this time into the other direction:

UK Composite PMI values were expanding in Q2 but UK GDP contracted by 0.7%.

It seems that the United States and the UK purchasing managers compare their situation to the previous more "desperate" years and therefore they are currently more positive than the GDP reality. Our conclusion is :

Markit Purchasing Manager Indices possess a high degree of sentiment and have to be taken into perspective against sentiments of previous years.

Markit Composite PMIs do not contain construction

By pure definition the Markit Composite PMIs takes into consideration only the output in manufacturing and services, but not construction. Interestingly the German construction PMI is also in contraction with 44.6, but the UK is expanding with 50.9. But also here negative sentiment depressed the German index:

"German constructors ... signaled concerns regarding the Euro zone debt crisis as well as the outlook for public spending. Furthermore, the degree of negative sentiment was the greatest since last December."

In March 2011 the situation was completely different, German construction PMI stood at 61.3. Similarly as the over-positive sentiment in 2011 today's negative sentiment seems to be exaggerated.

PMIs do not reflect trade balances or current accounts

Portfolio investments are flowing out of the euro zone as reflected in the following graph from the ECB.

According to the Asset Market Model these flows weaken the euro but also the construction PMIs. The negative current account of the euro zone in 2011 has become positive in 2012 thanks to the by 15% devalued common European currency. Employment in the core euro states has remained more or less stable. Provided that global demand does not become weaker, we expect the German and Swiss trade surpluses to rise further at the cost of American and British producers. This, however, is not sufficient to justify current overvalued levels of German stocks.

Investment recommendation and summary

In summary, our research has shown that the Markit Purchasing Manager Indices exaggerate the current negative sentiment in Germany and underestimated the German growth by 0.8%. At the same time Markit overestimates the UK GDP by more than 0.8%.

The apparent shortcoming is that PMIs do not reflect trade balances, current accounts or government spending. The Markit composite PMI does not question construction purchasing managers. Therefore PMIs are only one of several leading indicators, but should never be used as sole indicator.

Thanks to reduced consumer spending and the low EUR/USD exchange rate, the European current account surplus should continue to expand. If global demand does not collapse the weak euro is a chance for Europe to regenerate. The Swiss National Bank (SNB) was one investor that considered that the EUR/USD exchange rate of around 1.23 was sufficiently low to increase their euro exposure versus other currencies.

We recommend long-term investors to be go long the euro (long URR) and the Swiss franc (long FXF) against USD and GBP (short FXB) at EUR/USD and of 1.2050 with the next wave of bad euro data or Greek exit fears. They should reserve enough capital to double the bet at levels of EUR/USD of 1.10, for the case that euro zone problems worsen.

We are sure that the euro will not fall under $1.10, because US interest rates and growth will be limited for years. Even if the reduced consumer spending in the periphery caused GDP contraction, we see the positive aspects in form of better trade balances especially for Italy. A euro zone collapse seems still far for us, unlike the exit scenario of one or more weak Euro zone members. Which again will strengthen the euro zone.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.