Practice Good Theory

The monthly U.S. unemployment report is not much use as a real-time statistic. The data is meaningful only after it has been revisited, revised and is no longer current. Two factors make it difficult to compare the report with the "normal": seasonality  and the birth of new businesses. Adjustments are required. Total Non-Farm Employment (No Seasonal Adjustment) Seasonality: Statisticians adjust monthly data for seasonality in order to make better comparisons. For example, every year total employment drops sharply from June to July ( see chart ). If it drops this year, it is not necessarily bad news -- it could be an expected seasonal pattern. Instead of reporting the actual ebb and flow, statisticians often report a seasonally-adjusted number. A drop that is much less than normal , may be reported as an increase in employment! Birth-death Adjustments: A second adjustment by the BLS is an attempt to estimate how many new businesses have been formed, but ar

QE3: The Fed just announced a larger-than-expected, and open-ended "QE3". Do not expect significantly higher price-rises in the medium term (at least the next few years). Definitely do not expect hyper-inflation.  Most people who predict hyperinflation use some variant of the "Linear Quantity Theory of Money" and the concept of a banking "Multiplier". Both these ideas are false (except in a fuzzy, informal way).  [ Ludwig von Mises criticized the Linear Quantity Theory of Money , but many Austrian-sympathizers still continue to apply it.] The typical interpretation of the Quantity Theory of Money takes the view that there are two important aggregates: on the one hand, there is money; and, on the other hand there are goods traded... with other factors staying mostly constant over the short-run. This model leads people to think: more money is being created, so prices of goods will go up. Money vs. Goods: There are issues with both sides of the Quan

Unemployment: Yesterday's jobs report (Sep 7th) was reported as negative, after last month's positive. The monthly fluctuations have become standard since 2010 (see graph). Basically, the growth in jobs has been flat -- around 125K jobs per month since 2010 . It takes about 150K jobs just to keep up with population growth, but people have been dropping out of the job-market. Consequently, the official unemployment rate has dropped slightly, and ever more slowly. GDP: Huge amounts of "fiscal stimulus" have brought GDP back over its pre-recession level. The piper will have to be paid some day... but not yet. However, if one subtracts "transfer payments" from GDP, or if we look at industrial production, or the total number of people employed, we are still about 97% of the pre-recession levels. ( See this August 5th post from the excellent Calculated Risk blog for details .) Retail Sales:  Though retail sales rebounded from its recessionary lows,