Manufacturing is among the most important industries in our economy. Although employment in manufacturing has been on the decline for several decades, the current condition of manufacturing can certainly still provide us with information about our economy presently. This is especially true with data pertaining to the manufacturing of durable goods. Durable goods are goods that do not wear out quickly. This would include cars, home appliances, and furniture to name a few. We would expect for durable good consumption and therefore production to slow when the overall economy is not doing well. This is because people tend to delay purchasing higher-priced durable goods when times get tough and people are forced to cut back on personal spending. As a result, companies usually react by producing less.
The Federal Reserve publishes the Industrial Production Index each month to measure the level of manufacturing in our country. This index has many subindexes that are used to track the progress of specific sectors of American manufacturing. To begin my analysis, I collected data from 12 of these subindexes of durable goods between September 1972 and December 2015. Then I calculated a moving linear slope for each set of data. By doing this I could see when manufacturing was increasing or decreasing since 1972:
This graph is informative but we can’t easily see how many of the 12 selected subindexes are increasing or decreasing at any given time. Knowing how many of the selected indexes are increasing or decreasing would provide us with another way to gauge whether a certain downturn is contained to a certain sector or not. A more widespread downturn could help us discern potential recessionary time periods. To investigate this I constructed the graph below which measures the number of subindexes experiencing a downward trend at a given time. Meanwhile, the gray line indicates actual recessions through 2014 as determined by the US National Bureau of Economic Research.
One thing to notice right away from this graph is that a majority of the 12 subindexes enter a downward trend just before or during US recessions. The current state of this sum is down to three, which would seem to imply that we aren’t necessarily in immediate threat of recession. However, there are reasons to be negative. The middle of last year featured a few months in which the number of downward trending subindexes reached 10. I have drawn a line in the above graph to make this more clear. While the indicator has since come back down, the fact that the indicator reached 10 at all is a weak sign. In fact, notice that just before the 1980s recessions the indicator climbed to 9 and then receded to 5 before spiking back up. Similarly, just before the 1990-91 recession the indicator climbed to 11, receded to 1, and then spiked back up to a maximum of 12. And finally, this pattern also took place prior to the Great Recession of 2008-09 when the indicator climbed to 9, receded to 3, and then spiked to 12.
I hypothesize that this indicator will spike back up in the coming months. One reason for this thinking is that there are reasons to be negative about American manufacturing. First, the number of new orders made by manufacturers has been very weak as of late. For example, if we examine the annual percentage change of the level of new orders for capital goods we see increasingly negative growth:
Another reason to be negative about American manufacturing is that the OECD’s manufacturing confidence indicator is now in negative territory. This indicator is based on monthly survey data and measures manufacturers’ confidence in future business. Below is a graph of the confidence indicator since 1972:
Since the recession of 2008-09 the confidence indicator has been positive for all but two months. However, at the end of 2015 the indicator once again ducked below zero, and this time I don’t see good reason for why this level of confidence should turn around. Corporate and worldwide weakness that is beginning to take hold should continue to prevail in my opinion. Therefore, I believe the level of confidence should remain subdued.
Overall, I gather that manufacturing seems to be weakening, and weak manufacturing is usually associated with recessions. Thus, my conclusion from this data is that chances of a recession soon are decently good. And remember, there are several reasons to believe this time around will be especially bad.
This Wednesday the Federal Reserve will release the January Industrial Production Index data which should offer a glimpse into the direction of manufacturing levels.