Among the 25 entries in the table for composite PMI values in August, only 6 show month-to-month increases in August. Those showing improvement are Italy, Sweden, India, Saudi Arabia, the UAE, and Egypt. Three of them are countries in the Middle East, India was another, Sweden, a northern European economy, and Italy, a Mediterranean European Monetary Union member, was the final member of this group. There's nothing special that stitches them together apart from the three Middle Eastern countries that are geographically concentrated and that benefit from the ongoing high oil prices in the world economy.
In July only seven countries had improved month-to-month and in June on the eight countries had improved month-to-month; for the most part these were different groups of countries although Russia improved in July and June, Saudi Arabia improved in August and June, Ghana improved in July and June, Egypt improved in August and July, all of the rest of the cases were isolated.
Similarly, over three months only nine countries have improved compared to six-months; none of them were among the group of the largest economies in the table. Over six months nine countries improved compared to 12-months and year-over-year 11 countries improved compared to the 12-month period earlier; that count still falling short of representing half of the group.
The percentile standings show that a large proportion of the August values still reside in the upper portion of the full range of values these countries have had in their PMI composites from January 2018 to date. However, if we look at the queue percentile standings that place the August value in a particular spot in its historic queue based upon the number of observations above and below it, rather than based upon the maximum and minimum values in the range, we get a very different result. That result reveals the month's readings to be quite a bit weaker than range standings suggest. Fully 16 of the 25 values in the table reside below their historic medians which means on a queue percentile basis they have rankings below their 50th percentile. Most of them, in fact, are far below their 50th percentile indicating extremely weak standings. Nine of the 25 members have percentile standings in the lower one-fifth up their historic queue of data. That means that they have been stronger than their August values 80% of the time or more. Only two countries have percentile standings in the top 10% of their queues: those are India and Saudi Arabia.
The composite – service sector plus manufacturing sector- PMI paints a picture of the global economy that is weakening and paints a picture of a global economy that is already weak. At the bottom of the table, average and median data for benchmark countries and aggregates provide more perspective. The unweighted average for the U.S., the U.K., the European Monetary Union, and Japan has a PMI value of 48.2 in August and 49.8 in July; both indicate that those countries on an unweighted average basis are showing contracting economies based on the averaged composite PMIs.
These are composite PMIs not just the manufacturing sector; they include the broad services sector and therefore they're broad gauge of these economies. PMI data can differ from ordinary economic data which I referred to as accounting data since GDP and traditional statistics tend to count the data rather than to deal with the sorts of breath statistics that PMI data are based on. However, the two data presentations do tend to give us many of the same signals and the PMI data based on breath are well known to be sensitive to changes in economic conditions; they can serve the role of a canary and a coal mine to warn us when something is starting to go wrong.
The unweighted average for the U.S., U.K., EMU and Japan readings progressed from 53.4 over 12 months to 52.6 over six months to 50.2 over three months that's a clear weakening. For the Bric excluding Russia, the pattern is different: 52.3 for the 12-month average, slipping to 52.2 for the six-month average, but then rising to 54.7 over three months. The average for the full 25-member group slips to 53.0 over six months from 53.6 over 12 months and again to 52.3 over three months. The median for the full set of countries also shows slippage from 53.8 over 12 months to 53.0 over six months and to 51.8 over three months. Both the median and the average measures for the full data set show slippage.
However, when it comes to contraction, August shows only 10 of 25 jurisdictions below 50 compared to 7 in July and 4 in June. Over three months only six averages show below 50 values compared to five over six months and 5 over 12 months. This is a starkly different comparison from slowing. Statistics on slowing show 19 jurisdictions slowing in August compared to 17 in July and 17 slowing in June compared to May. Over three months 16 jurisdictions slow compared to six-months, and in six-months 16 slow compared to their 12-month values. Over 12-months, however, only three jurisdictions slow compared to 12-months earlier. Slowing is extremely widespread; but contraction is still not very widespread, occurring in its most frequent month of August in only 10 of 25 of these reporting jurisdictions – still that's 40%.