The market had a field day with the DOW dropping 370 points yesterday. All of this because the Institute of Supply Management released the non-manufacturing (services) report indicating a contraction in the service sector’s business activity to 41.9% in January from 54.4% in December.
My first gut reaction was quite a shock. When you look at magnitude of such a drop you wonder if we really are heading into a recession. Maybe we are in one already. But what does 41.9% really mean? Are the investment powerhouses reacting off the gun, selling off their positions, protect their assets, and asking questions later? What does 54.4% really mean? What specific areas of the service sector contributes to this number and how is it calculated?
Having knowledge of such economic factors can help guide investment decisions, so I thought I’d try to break out the meaning of this index over lunch today. I wanted to see if there was any way to utilize the report to optimize my virtual portfolio for the rest of the year. First, we need to break down the ISM service index.
Non-Manufacturing
The report attempts to gauge how the service sector of the US economy is doing. For example, it will cover how the hotels and restaurants are doing as opposed to how Ford is doing. The Institute of Supply Chain Management sends out surveys every month to these businesses to answer a few simple questions. The survey isn’t just sent out to anyone, but directed towards the people who have the power to buy stuff and hire people. If you aggregate enough of this information you should be able to get a pretty good gauge of the service sector’s health.
What does the Non-Manufacturing Index try to measure?
The index is really comprised of 4 components and tries to measure if the overall service sector is contracting, staying the same, or expanding. The ISM just started doing the composite version of the formula in January 2008 to most likely get a better picture of everything. The 4 components are:
- Business Activity (general business condition)
- Employment (hiring more, freezing, and layoffs… I hope not)
- New Orders (are you buying more stuff, or less?)
- Supplier Deliveries (how is fast are your suppliers delivering stuff?)
In each of these components, the respondents must answer one of the following about what changed regarding a given set of criteria. The 3 answers to choose from are Better, Same, or Worse. Example:
What you end up with is an allocation of answers into those 3 buckets for each of the 4 main components mentioned above. Each of the 4 components will have a series of questions. Let’s take the Employment category for example. Let’s say one of the questions was posed:
At first glance what does that look like to you? It’s more weighted on the worse side of the fence, so sentiment amongst purchasing executives in the service sector see employment activity worsening. The ISM then uses what is called a “diffusion index” to make the number easier to digest. When you go month to month, the breakdown amongst the 3 buckets will vary. It’s harder to gauge how bad or good something is doing. So using a “diffusion index” pretty much generates a single number to streamline it a bit. The formula takes the “Better” component and adds it to one-half of the “Same” component. So using the example we have above, the index value for this question would be:
6% + (70% divided by 2) = 41%
Now let’s try another scenario:
What does this look like now? It looks like nothing has changed really as 5% of respondents think things are getting better, and 5% think it’s getting worse. The remaining 90% see nothing changing. Running this through the formula you get:
5% + (90% divided by 2) = 50%
And that’s the simple beauty of a “diffusion index”. 50% means nothing changed. If the number dips below 50%, the survey respondents overall think things will move in the “worse” direction. If the number moves above 50% then overall sentiment moves towards the “better” direction.
So what ISM does is average out all these index values for all the questions in the survey and adjusts it for seasonality changes. The resulting Employee Activity index value for January was 43.9%.
The overall measure is to equally weight the 4 components (just average them out). This is what you get:
What am I supposed to do with this data?
Everyone is going to interpret this differently but it sure seems like everyone was selling yesterday on the news. There maybe an opportunity when everyone is moving like a charging herd in one direction. Digging further into the ISM report I extrapolated some nuggets of good info.
In each of the 4 main components, there are some areas of the service sector that actually show positive growth. Now let’s say you take any given stock screen and then apply another filter to weed out those companies that belong in those positive growth areas identified by the ISM report. Maybe those sectors are recession proof or are benefiting from weak US dollar through foreign exports?
Let’s see which sectors experienced growth despite the downbeat message (the winners) [1]:
- Business Activity : Utilities and Educational Services
- New Orders: Utilities; Professional, Scientific & Technical Services; Educational Services; and Health Care & Social Assistance
- Employment Activity: Transportation & Warehousing
- Supplier Deliveries: Management of Companies & Support Services; Wholesale Trade; Educational Services; Finance & Insurance; Public Administration; and Information.
At first glance a few service sectors seem to have a pattern, particularly Educational Services which show up in 3 out of the 4 main components that showed growth. You could take some of this data and scan for those well managed companies that belong in the educational services sector which have also experienced a sell off. If next month’s report demonstrates the same pattern, and the market continues to beat down companies that generate good cash flow, it will provide a good base for my next set picks.
The report contains a lot of interesting data, yet I wonder how many people actually read it. Ready, Fire, Aim.
