Identifying “Hard” Data

During recent weeks, much of the punditry has suggested a sharp difference between “hard” and “soft” data.  The former seems to show sluggish economic growth while the latter paints a more optimistic picture.  The general distinction seems to be that surveys are “soft” and other data are “hard.”   Results are rather arbitrarily placed in one group or the other.

Test Your Hard Data IQ?

Since reliance on surveys seems to be the acid test, let’s try a little quiz.  Which of the following reports use surveys, and which do not?

  • Payroll jobs
  • ISM manufacturing index
  • Wholesale inventories
  • Retail sales
  • Unemployment rate
  • Labor participation rate
  • PPI
  • ISM non-manufacturing index
  • Consumer confidence or sentiment
  • Business optimism
  • Durable goods orders
  • New home sales
  • CPI
  • Building permits
  • Personal income and spending
  • The decennial census of U.S. population
  • Existing home sales
  • JOLTS report
  • Business inventories
  • Housing starts
  • Regional Fed indexes – Chicago, Empire State, Philly Fed, Dallas, etc.

Before going on, please make sure that you have indicated “survey” or “non-survey” for each of the reports above.

Criticism of Survey Data

Most of the survey critics dismiss such data as “soft” and undependable.  This is often an assertion that surveys always involve speculation about future behavior.  Without exception, the survey critics do not include anyone who has actually designed or administered a survey.  A real expert would know that most of the complaints are treated on day one of the methods class.

Please consider this obvious example.  Suppose we ask someone, a month before an election, for whom they intend to vote.  Now suppose we ask in an exit poll.  The respondent might still give an inaccurate answer in the latter, but logic suggests it will be more accurate than the first.  Logic is confirmed by results.

Try this example.  We ask a purchasing manager if he/she is ordering more this month than last.  Or if prices paid were higher?  It does not involve speculation – just an honest report of facts.  Suppose we ask a Chinese purchasing manager the same question.  Are we just as confident of the answer?

Many questions are carefully worded to make it easy to give an honest answer.  It is a technique taught in the classes.  Most of the critics have never looked at the actual underlying surveys or considered the issues.

Quiz Answer

All the reports listed above involve the use of surveys.  All of them.

Many pundits pick and choose data to support their viewpoints.  The “hard versus soft” meme is the latest such effort.  Here is a test.  The choice of classification has been very loose and arbitrary.  Despite this, it has been taken seriously by nearly everyone.

We have seen many new fans of the Atlanta GDP Now tracking.  It showed weak growth in the first quarter.  During the second quarter the early returns are 3.5% to 4%.  Many sources will react by finding a new favorite indicator.

Investment Conclusion

The economy has a better footing than most sources allege.

Your decisions will be better if you rely upon sources that are intellectually honest in the consistent use of data.

You may also like


  • Sasha Naryshkine May 10, 2017  

    Hi Jeff.
    Thanks for the excellent and continued blogging, all of which are wonderfully rich and entertaining. To answer your questions and to be direct (as is possible), I use these data-points to “see” where one is. It makes for little use however when picking a business to buy as a long term investment. The level of the market could be whatever you want it to be, the level of the stock relative to the company earnings are all that matters. You can be wrong on a stock for a number of years, that may well represent a wonderful opportunity in the interim (i.e. you can own more and more of the same business at a lower or equivalent price). Whilst these are some sort of litmus test as to the health of a business, they help little in determining what is a very good investment, and what may be a good price.

  • Robert Simmons May 10, 2017  

    How does the conclusion follow from everything before it?
    You’re right that some people are too dismissive of soft data, and that all of it is based on surveys. And that there are gray areas. As Sumner says, market forecasts > hard data > soft data > nothing.

    • oldprof May 10, 2017  

      Robert — Sometimes a conclusion just emphasizes what is important about a post. It is not always a syllogism. I can’t really do a comprehensive economic review in every post, but I cover it pretty well in WTWA.

      Those who start with a mission about the economy are always uncovering something. This one quickly pushed into the mainstream with little real analysis. It is easy to do since it plays upon what people already think.

      I have high regard for Scott, and I’ll look for that comment. My observation is that people in general are too dismissive of methods where they have no experience. Note the absence from this financial media debate of those skilled in survey research.

      Thanks for an interesting and provocative comment.