Manufacturing PMIs Show Some Improvement But Amid Clear Ongoing Weakness

Month-to-month manufacturing PMI changes were mixed in April with increases in eight of 18 observations (two unchanged by assumption because of missing data). Except for the U.S., the largest economies generally worsened in the month (Euro Area, Germany, France, the U.K, and China).

The progression to better (or less bad) conditions is clearer looking at three-month changes. The three-month averages of the manufacturing PMIs show weakening compared to 6-months in only 5 of 18 categories. Over six months things shift again, and PMIs are better in only 6 of 18 categories. Over 12 months, this trend continues as only 5 of 18 are better.

These metrics underscore that the current progression to ‘better’ (...or not as bad) is relatively recent and that the concept of ‘better’ applies just to comparisons over very recent months since over 6 months and 12-month conditions broadly are worsening.

That is hardly surprising… When we turn to engage with the column on rank or queue standings of the level of diffusion readings, weakness is the overpowering result. The median standing is a 24-percentile standing; that places the median for the group in the bottom 25 percentile of all observations since January 2019 – that is an extremely weak median.

One version of this month’s data is that there is some sort of revival going on… another version is that… “Well, yes, things are better, but not by much.” I am much more in the second camp than in the first camp. Still, it is notable that central banks have been hiking rates and inflation remains far too strong in most countries/regions and yet there has been some improvement in economic activity. Even if it is a minor effect, it is contrary to expectations and for that reason still notable.

The median manufacturing PMI value for each of the last three months as well as each of the three sequential periods referred to above, the PMI medians all are below 50 – indicating that contraction is most common.

The distributional data on the manufacturing PMIs also underscores that conditions are still broadly weak. Since the PMI data in Table 1 are just country by country (or by region) and since we do not apply any weighting scheme to pool the data, the distributional features are another important way to gain context on what the PMI values mean. In table 2 (below), we see that one-third of the observations are in the very moderate growth range of 50 to 55. Only 5.6% of observations are at diffusion values that are higher, in the range of 55 to 60. On the other hand, 61% of observations are in the moderately lower 40 to 50 percentile range in April. In terms of the month-to-month comparison, the monthly improvement has shifted the 50 to 55 cohort, up by nearly 6 percentage points and has reduced the sub-par 40-50 percentile grouping by the same proportion. However, compared to February, there is still a large proportion of reporters in the sub-grouping of 40 to 50 in April.

Over three months compared to six months, we can also identify a similar identical proportional reduction in the sub optimal range flipped near 6% of reporters into the moderate module at 50-55. But compared to 12-months, the 40 to 50 cohort is proportionately larger over three months.

Anyway, you array the data, the answers are the same- that’s a good thing: these are robust results. There is some recent one-month to three-month improvement, but it is small. The overriding assessment of April conditions based on the levels of the diffusion data being reported is that they are weak. Moreover, with so many problems in circulation from inflation to war between Russia and Ukraine to ongoing supply chain reductions and inflation-fighting, the prospects for this improvement to play out in the future seems dim. It is more likely we are in the twilight of the effects of Covid era stimulus and with policy over the past year more engaged on the tightening side, it is far more likely that effects of more tightening are in the pipeline and ready to play out across reported data. That’s a reason for central banks to be cautions with policy especially with banking sector issues having emerged as a separate potential worry.