Cycles in consumption expenditures

Why should we expect cycles to be part of macroeconomic dynamics? In short because spending increases often change cyclically, in investment and consumption in particular. Today let’s start with just looking at consumption. As a consumer, if the hottest new product arrives in the market, what will that likely do to a firm’s sales? Clearly there will be boost in spending as consumers get the information about the new products and decide to upgrade their technology or even the “look” of their wardrobe. Let’s take a few examples just to make this more concrete.


First, think about flat screen TVs: when they first came out most of them were relatively small at around 32”-40” and operating at 720mHz dynamic resolution, but many of us decided that they were a marked improvement over a regular TV, so decided to upgrade.  After a year or so, a new version of the flat screen came out with a 1080mHz dynamic resolution, so once again many of us decided to upgrade or to replace another of our non-flat screen TVs. Then they got (much) bigger and picture quality has improved as well, and once again many consumers changed out their TVs and upgraded to the better TV. If even a small number of consumers do this there will clearly be “growth cycles” created in the sales of these types of products, as spending will swell as this new technology hits the market. These cycles are clearly going to be irregular and will depend on the frequency at which breakthroughs in new technology occur, and how stable consumers expect the technology to be.

Second, think about non-iron shirts. I bought my first non-iron shirt about a year ago, and have been so thoroughly impressed, I’ve decided to upgrade my entire wardrobe. Now in the retail clothing sector there are clearly 4 seasons in the year, so we would expect there to be cycles on a quarterly basis in the data. But of course the data we prefer to use in economics is “seasonally adjusted”, meaning that systematic seasonal variations are adjusted for. But what happens if these variations are not systematic so that some seasons seem to do better than others. Clearly if this happens we can see cycles being created – indeed the season in which non-iron shirts were heavily marketed to men was a season where I spent much more on clothes than usual, and likely others did too, creating a spending cycle which the "seasonal adjustment" will not properly take into account.

Lastly, let’s think about vehicle sales. Many vehicles have not been replaced by owners recently because of the economic downturn, so apart from when the “cash for clunkers” program was in place, the general public has not been replacing vehicles. So when economic growth recovers, there will be a delayed demand to replace vehicles, and then there will be a larger number of vehicles of a certain vintage on the roads. Clearly they will likely be replaced at similar times in the future, leading to a bulge in expenditures at some point in the future, but the point is that these expenditures are likely to be more synchronized than they would have been as they were all purchased at a similar time.

Hopefully you've been convinced so far.  Next time I'll look at how some of the techniques I've been using in my own research break down consumption expenditures into cycles at different frequencies.