Practical Economics: Squishy Numbers, Economic Indicators & the Mainstream Media

Practical Economics:  Squishy Numbers, Economic Indicators & the Mainstream Media 

Previous Student: 

Actually, I suspect the biggest unknown variable in every model of the dynamic economic forces at work that does not lend itself well to hard data measurement is – Consumer Confidence. We try to put a number on that every month, but is assigning a number truly reflective of reality?

Comments Regarding Soft Data

I wholeheartedly agree. We try to take all sorts of measurements to predict the economy, but I really think Consumer Confidence might be the most important. If people are confident that they will stay employed, that they will have steady income, then they will be more likely to spend or invest money. Yet measuring this confidence is difficult at best.

One important piece of the puzzle, as far as making meaningful economic predictions, is knowing something about consumer sentiment or Consumer Confidence. But Consumer Confidence cannot be measured objectively, period.

What we use is something the University of Michigan has developed called the Consumer Confidence Index. Every month, a certain number of consumers are surveyed and asked similar questions. Their answers are tallied and the end result compared to a base of 100 developed in 1982 or thereabouts. However, no matter how much mathematical care we exert, the monthly Consumer Confidence number is still soft data.

Squishy Numbers

The Consumer Confidence numbers are ‘squishy’ because they are subjective. But Consumer Confidence numbers are measured every month. That allows for some good comparisons, month to month, which removes some of their squishy nature stiffening up those numbers. However, there are also wide swings (cause unknown) from month to month that are more related to ‘mood swings’, than anything else. Remember though, one’s current mood has much to do with one’s current purchases.

The Consumer Confidence number is soft data, while many of the other economic indicators are hard data. The difference is that hard data numbers are counted, like totaling out-of-pocket dollars used or noting the time spent, while soft data numbers are completely subjective, that is, estimated by each person, subjectively, one-by-one.

For instance, soft data numbers are student survey numbers generated at the end of each university class. If you are asked, “on a scale of 0 to 5, zero being ‘never’, 5 being ‘always’, how do you feel about this?”, then the response you give is soft dataSoft data is an attempt to measure feelings or happiness or satisfaction or subjective value, etc.

Measuring feelings correctly is hard to do, but not impossible. At least few in our society thinks such measurement is impossible. However, and this is the critical issue, measuring feelings can be done well or be done poorly. Usually, it is not immediately obvious if the feeling measurement has been done well or not.

‘Hard Data’ and ‘Soft Data’ are NOT similar

Whereas the number of Americans unemployed at a given time can be counted (then estimated, but we assume the approximation is close and statistics tell us it is) by anyone because hard data is not feeling, but something that is counted, with all the attendant problems of counting. Soft data is never really counted per se as it is non-tangible by definition. Instead, because soft data is an attempt to measure feelings, we first try to identify which feelings we want to measure, then we put a number on those feelings, and then those numbers are tabulated.

There is a huge difference between hard data and soft data. I contend that soft data is almost always more important, but always much more difficult to measure. The reason the University of Michigan survey on Consumer Confidence is important is because the survey is done carefully and the same questions are asked over and over again, every month and every year.

The rich irony is often soft data is more important to business success than hard data. We recognize that fact when we say a person’s gut instinct is great. Or we identify that her intuition is almost always right on. Do employees happily anticipate going to work every day or do they dread it? Do you enjoy talking to your boss or do you try to avoid that person whenever possible? Examples of soft data in the workplace abound.

Yet actual difference between hard data and soft data illustrates one of my complaints with the academic study of economics. Some economists pretend that since economics can be subject to rigorous mathematical and statistical analysis, therefore, if we always follow the MC = MR rule (or some other economic principle) under every circumstance, then we will never be wrong. If only life were so simple.

One of the main reasons that economists have many different opinions while using the same information is because many crucial measurements in economics stem from soft data, like Consumer Confidence, which is decidedly NOT hard data.

Mainstream Media Problem

The worst part of this issue is how the mainstream media treats economic soft data and thus amplifies the reality problem. On a non-stop, 24/7 basis, the economically-challenged mainstream media inundates Americans with soft data masquerading as hard data. In fact, I am sure that most people in the economically-challenged mainstream media have little idea there is a difference. And perhaps those that do know better choose to bask in the power of broadcasting misinformation.

For instance, on a daily basis, the value of the overall stock market moves up or down based on the new information coming forward that day. Truly the stock market considers  many squishy numbers and issues daily, then the stock market translates both the soft data and the hard data into only hard data — reflected by a new stock price. But no one besides Warren Buffett seems to have a clue what the exact relationship is. And there are thousands of public companies trading millions of shares every day.

No one has ever been able to discern exactly why stock values move up and down. Yet often a single slice of the daily news (soft data – e.g. the Greek debt situation worsened) is identified by this or that analyst, then that analyst’ opinion is picked up by the economically-challenged mainstream media and confidently deemed the sole culprit causing the up or down movement in the stock price (hard data). This economically unreasonable scenario happens every day.

Likewise, we are constantly bombarded by the economically-challenged mainstream media claiming this soft data poll (usually ignoring relevant statistical issues) to be hard facts or that soft data survey to represent reality, whereas the truth is that what we are being given is tabulated feelings dressed up as hard data. Most people recognize the difference between soft data and hard data, the importance of soft data and the difficulty measuring soft data, but the mainstream media sometimes seems clueless.


May I conclude with a soft data analysis of my own? “On a scale of 0 to 5, zero being ‘not at all’, 5 being ‘perfectly’, how well do you feel the mainstream media is doing at disseminating meaningful information? A generous score from me on this scale would be 0.5, or perhaps 10% of the time, a scarcity largely caused by mainstream media confusion between soft data and hard data.