Watch Now


Biased against sentiment

PHOTO COURTESY OF SHUTTERSTOCK

We all have biases.  It is part of our basic human nature. There is nothing inherently wrong with having a bias or a preconceived notion. Sometimes biases are incredibly useful; they prove to be very valuable. For instance, I admit to having a very strong bias against investing in asset-intensive companies, with high levels of debt, low margins and low returns on capital, yet they are selling for high multiples of valuation. If there is a danger in having a bias it often stems from either not recognizing it as a bias, or not even realizing you have a bias, or by practicing that all too common human frailty of ‘confirmation bias’ – i.e., looking for evidence that agrees with your preformed conclusion or thesis and ignoring or failing to critique information that does not fit your thesis or contradicts your preformed conclusion. 

That said, i openly admit it. I also have a bias against – actually detest – most surveys based on ‘sentiment.’ Asking “How do you feel?” does not usually produce reliable data that can be used to predict behavior, or micro-economic activity, much less macro-economic trends.

I will spare readers a full statistical study of how useless sentiment surveys in general can be, after I point out the problems with what I consider to be one of the most over-rated, over-talked about, over-reported and useless sentiment surveys – consumer sentiment. Most find it surprising, but back-tested against what you’d think it should be indicative of, consumer spending, it has little to no correlation, and no predictive value (unless you flip it, delay it by six months and try to use consumer spending to predict consumer sentiment, but I digress). 

Why doesn’t consumer sentiment predict consumer spending? Simple – consumers don’t spend confidence, they spend income (expansions and contractions in consumer income are highly correlated with expansions and contractions in consumer spending). Consumers can be wildly optimistic and feel great about life, but if their income goes down, so will their spending.

Consumers can be horribly pessimistic and feel lousy about life, but if their income goes up, so will their spending. For the cynics out there, who may use this as an opportunity to make a statement about the low savings rate of most U.S. consumers, I will point out that even if you routinely saved 50 percent of your income, your spending would still go up and down in lockstep with the amount of income you have left to spend after putting 50 percent of it into savings.

 PHOTO COURTESY OF SHUTTERSTOCK
PHOTO COURTESY OF SHUTTERSTOCK

After this long, drawn-out confession, you might be surprised to learn that there is one sentiment survey that I do hold in high esteem – the ISM (Institute for Supply Management, or what used to be widely known as the PMI – Purchasing Managers Index). 

Why does the question of “How do you feel?” matter in this instance? To be fair, the survey does ask very specific questions that are related to metrics for which the responders are professionally trained (which helps produce more reliable data), but the essence of it is basically this – it reflects whether or not they are signing or seeking to sign purchase orders. Do they need to sign more of them or fewer of them? That is what the ISM seeks to determine. It queries on factors that have been historically proven to determine current and future activity. For years, I have had numerous trucking company executives and managers confirm that they share my view. Why? Because one of the single strongest indicators of freight demand, at least for the near future, is the ISM. 

What is the most recent ISM suggesting? For FreightWaves SONAR users, it is as simple as typing in ISM.PMI and then adjusting the scale and time-frame to meet your needs.

Are there any issues? Any potential pitfalls to using the ISM as one of your key diagnostics in assessing how strong or weak demand will be in coming weeks? Yes. Although it has proven to be highly useful and reliable in predicting freight demand – especially truck freight demand – in the upcoming days and weeks, it does have a limited duration value.

That is to say, the ISM may be very insightful about the current and “coming soon” behavior of purchasing managers, and the freight volume that occurs as a result, but it does not predict what they will be doing six months from now. Conditions change, and both customers and company managements change their minds. Not signing purchase orders today doesn’t mean they won’t be signing as many as possible six months to a year from now; just as writer’s cramp from signing purchase orders today doesn’t mean purchasing managers won’t be playing solitaire at their desks in six months. 

What about today? The current ISM level of 55 reflects a solid demand environment, but is nowhere near as strong as it was last year when it was routinely above 58 and even exceeded 60. Sound familiar? I suspect it does. Upon reflection, I suspect that most shippers, trucking companies and freight brokers would say they “felt good” about the ISM’s ability to reflect current conditions and even predict conditions for the next few weeks, which is a sentiment without an index that we can all have confidence in.