A casual observer of news about economic indicators has more than enough reason to be puzzled.  We hear news of soaring stock markets alongside dropping home prices, continued joblessness and fears of inflation.  These indicators suggest a significant level of cognitive dissonance about the economy.  The truth is there are three things about economic data that oftentimes makes them appear bizarre—the timing of its interconnections, the quality of the data, and misunderstood randomness.  Let me explain. 

First, the different parts of the economy work together, but at often long and uneven lags.  The economy can only grow in the long run with more folks working.  In the short run, it can grow at a furious pace with high unemployment.  As new factories start up, existing workers move from part time to full time, and as businesses place orders for new inventory, we can see rapid expansion and high joblessness at the same time.  This makes it look like there is a disconnect between Wall Street and Main Street.  There is not.  Over the long run, stock brokers thirst for low unemployment, high rates of savings, high wages and productivity growth – just like Main Street.

Second, the data isn’t great.  The most breathless news stories are usually based on preliminary releases by federal agencies.  Virtually all the preliminary releases are later revised as new data become available.  So whatever you think about the jobless rate or new hires today is almost certainly wrong, even if the data release is the best available.  This is especially true as the area of study gets smaller.  It is simply a matter of averages.  The larger the area, the more likely the mistakes cancel each other out.  County data is worse than state data, and much worse than federal data.  After about six months, the numbers get pretty solid.  

The third and most problematic of the reasons for confusing economic indicators is the quest to assert  causation where none exists.  The truth is much of the daily or weekly changes in economic data from stock prices to employment numbers means nothing.  The stock market, which is subject to falsely-serious analysis each evening, is the best example.  To be sure, some things, like international unrest, can be linked to stock shifts, but most of the daily swings are pure randomness.   Even after tens of thousands of researchers with strong profit motive have spent a century studying stock behavior, no good model exists.  Indeed, the best way to predict tomorrow’s stock price is simple.  Start by filling a hat with slips of paper upon which are written the daily change from each of the past hundred days.  Then, choose one and add it to today’s stock price.  As hard as it is to believe, it is true and the proof is in the pudding.  If anyone could actually predict stock prices effectively he/she would be the richest person on the planet by the end of the week.  So, is it any wonder the stories about the economy reporters tell are so confusing?