Manufacturing Is Doing Much Worse Than The PMIs Indicate

Summary

• There's a trick in the PMI numbers which tells us that manufacturing is doing worse than the headline numbers tell us.
• The market more generally seems to be believing those headline numbers, this being something of an error.
• This factor does not apply to the services numbers which we should therefore take as being more accurate or indicative.

Yes, it's a recession

Clearly and obviously we're in a recession. But what we'd like to know is how deep is it? For that will at least start us down the track to working out how many companies are going to survive and also how steep the hill we've got to climb out of it is going to be.

We'd also like to get a heads up on this. Rather than waiting until a month after the time period concerned, as we would have to do with GDP figures and the like, we'd like to have some idea of what's going on now which tells us about the future

PMIs

This of course is where the purchasing managers' indices come in. For the US we have two versions, one from the Institute for Supply Management, the other from IHS Markit. They use slightly different panel constructions but we're not worried about their differences at this level. It's also not quite true to say that they're forward indicators - they are measures of what is happening now. Rather, they're measuring - what is being bought to make things out of - the things which give us a guide to what future production is going to be.

However, we have a slight problem with the manufacturing PMIs as they're not telling us quite what we think they are. They're showing that the world is quite a bit better than it truly is.

Suppliers' delivery times

This is something that is common to both measures, ISM and IHS. If supplier delivery times stretch out, then this is counted as being evidence of the economy expanding. Or of not contracting if you prefer.

The intuition is simple enough.

(Suppliers' delivery time indices from IHS Markit)

If everyone's getting busier, then there will be more call on suppliers for supplies. Which may well not be there and have to be produced. So, a lengthening in supply times could - should - be taken as an increase in production. In normal times that is.

Except, of course, this isn't normal times. Here the thought is that supply times are extending because the supply chain is broken. It's more difficult to get things when part of the supply chain is closed down. This is thus a sign of decreased, not increased, production.

(Components of PMI from IHS Markit)

There we can see the oddity of that one measure. Supply times are rising, indicating increased production by our usual measure. And yet everything else is diving. It's our intuition about supply times that is wrong here.

That is, there's an inversion in a part of the PMI and that means that it is underestimating how much manufacturing production is falling.

Services

As it happens we don't measure supply time when calculating the services PMI:

The latest data signaled a substantial decline in business activity across the U.S. service sector in April, as the COVID-19 outbreak escalated and emergency public health measures intensified. The rate of contraction accelerated to the fastest on record as client demand slumped and many businesses closed temporarily. New order inflows fell significantly as customers postponed or cancelled orders amid ongoing global lockdowns. Subsequently, expectations for the year ahead sank to their most pessimistic in the series history. Uncertainty and a further reduction in confidence led to the steepest decrease in workforce numbers on record. In an effort to retain clients, firms passed lower costs on to clients through the fastest decrease in output charges in the series history

Note that among the things measured supply time isn't there. This makes sense with services as we very rarely do wait months or even weeks for them to be supplied.

What this means

Currently the market seems to be reading the manufacturing PMIs as they would in more normal times. That is, underestimating the fall in manufacturing output that's about to be recorded.

It's still true that services are suffering much more than manufacturing but that's to be expected given the upfront, person to person, nature of much services provision. The gap is smaller than we've been recording though.

My view

I'm generally - and have said so here many a time - of the view that we pay too much attention to manufacturing output numbers. The entire sector is some 10% or so of the economy so even major changes here have limited effects on GDP. Services are 80% and thus hugely more important as a measure of the economy.

It's only the manufacturing numbers which the PMIs are overstating here, not services. Thus, we're not on the cusp of some large amount of misinformation hitting market prices. Only a small amount.

The investor view

The GDP numbers, when they arrive, are going to be more gloomy than the PMIs indicate. Or, if you prefer, the PMIs are wandering a little more than they usually do from outturn. We now know about this and so will be less shocked when those GDP numbers appear. But it is useful to be forewarned.

Manufacturing is suffering worse than we thought. A rough guide to how much by, perhaps 5 points on that 0 to 100 scale the PMIs use.

As to the larger lesson here, just don't lend too much weight to any one statistic unless you're very sure of how it is compiled and why. For the devil is always in the detail.

Analyst’s Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Comments

[Grain-of-salt]

Where do we go from here? hopefully not into another great depression.

Models show that if 80 percent of people wear masks that are 60 percent effective, we can get to an effective R0 of less than one. That’s enough to halt the spread of the disease.

We overreacted with the lockdown, which caused this recession and increased the national debt. We need to carefully reopen the economy and return to activity and productivity, do what we can to help the economy recover.

[Yanks56]

This is criminal behavior and he should be tried as such

[enigma22f]

The economy will be slow for a long time. When people don't feel safe, they won't go out and spend. Trump has undermined confidence by giving up on the virus and driving the country to reopen. It will backfire. We can't lock down forever but we can't pretend the virus doesn't exist, either. No answers, no confidence, no economic boom. But the market is forward looking and we will see better times...the question is when.

[Grain-of-salt]

It's up to governors, not the president.
Either way, continuing the lockdown for more than a few weeks is also unsustainable in many ways.

Look at the Covid numbers by age group. We need to isolate and protect the elderly and the infirm, maintain reasonable social distancing, wear a mask, wash our hands etc. But find the strength and courage to continue activity and productivity so we'll have the resources to cope with the challenges ahead