The following was posted as a comment on Paul Krugman’s blog on distrust of data.
I am one of those who distrust economic data because I see two problems with it. One is that I do not always agree with the economic theory upon which data collection is based and the other is the difficulty in accurately and totally counting economic activity.
There are a great variety of economic theories and date collection represents the theory of the collectors which may or not be correct. For example, money is usually defined as currency in ciruclation and bank deposits with several definitions depending upon what deposits are included. I prefer to think of money as a tool to facilitate the exchange of goods and services. How does one count or measure a tool?
Counting economic activity is another problem because it can only be recorded if there is a monetary exchange. How do you measure housework or the many small things we do for each other.
When I worked as a journalist I realized there are two types of figures. One we photograph and use to sell newspapers and the other put things into perspective.