Recently I discussed what a bubble (like the housing bubble) is, and how they behave. One thing that I noted is that bubbles need a way to start; a way for the band wagon to get going. To examine the genesis of the housing crisis in this post I will look at housing price data to see the first drastic upward movement in housing prices, examine the time period in which that first happened, and speculate on the origins of the crisis.
Identifying the Time Period: The most widely-used source of housing price information is the Case-Shiller Home Price Index. The Case-Shiller is a “constant-quality” measure of housing price, meaning if the index is independent of housing type trends (i.e. it isn’t influenced by the proliferation of “McMansions”). The Case-Shiller did not officially exist until the late ’80s, but Prof. Robert Shiller extrapolated backwards based on historical data, so I’m going to use that set. It should be noted that the numbers will differ from the actual Case-Shiller as they are on a slightly different scale, but the relative changes are exactly the same (it’s an index after all).
Looking at this graph, it certainly suggests that the bubble started in the mid-’90s. Note that housing prices spend most of their recorded history in the same range. There is a bit of a depression in prices post-WWI /Flu Pandemic/Great Depression, but the price of a normal house doesn’t stray much from where it has always been… until the mid-’90s. The acceleration in prices is clear by 1997, which we can assume means that the precipitating conditions started a year or two prior to that (i.e. 1995 or 1996).
Economic Factors in 1995+: As I’ve stated previously, the increase in housing prices does not appear to have been dictated by some massive change in what a house was. Houses are still houses, and no massive change was made to their nature. Instead, we look to the basics of supply and demand for answers:
Supply: Prices rise if supply falls.
- Were there fewer houses made in the mid-’90s? No, the opposite was true to some extent. The housing industry started producing more houses, albeit at a very slow pace. It is not a nimble industry.
- Another possibility is that material costs rose significantly. There was some rise in materials costs (graph below, again from Shiller’s data and on the same scale), but it was not very significant at all, and certainly not similar to the increase in price.
Demand: Prices rise if demand rises.
- Abnormal population growth that outpaced the ability of suppliers to grow would increase prices. However, there was no unique event like a baby boom is obvious during this time.
- Overcrowding would lead to increased prices. It could be argued that urbanization caused some rise. If everyone wants to live in the same place (cities), the price of those houses will rise. However, there was no monumental shift.
- Irrational and emotional purchases could occur. I would argue, however, that a significant portion of the population doesn’t suddenly act irrational without incentive, and this certainly would not drive massive growth in prices (at least not initially).
It doesn’t seem like anything was out of the ordinary in the supply/demand picture. However, there was a massive demand change that was occurring behind the scenes. To see this, we don’t need to look at the population growth, but the growth in the percent of the population that owned homes.
Why It Drove Prices: Ok, so a few percent of the population were added to the market. It’s only about 0.5% of households per year, so what? Well, that small percentage amounts to approximately half a million new households looking to buy per year that weren’t before. Now consider that in the typical year, there are approximately 200,000 new houses built. 700,000 buyers were being added annually to compete for only 200,000 additional houses. That’s a bidding war. The housing industry is not a nimble one that could accommodate all the new demand. While it was able to scale up to some extent, the highest they got was 315k new houses completed in a year, which was right before the bottom fell out.
So, it seems a reasonable conclusion that genesis of the US housing bubble was some cultural or policy shift that occurred around 1995 that caused housing ownership to spike. This spike set off what is essentially a bidding war. Admittedly, people (and banks) eventually did crazy things to help perpetuate the bubble, but that is what people do. In crazy situations, people do crazy things. It doesn’t change their culpability, but it also isn’t some new condition of greed and evil. It’s human nature.
That of course, leaves us with the question, what the heck happened in 1995? However, the gist of it is that the government did something very well-intentioned that worked out terribly, as detailed in another post.
[Note: Shiller data available at Robert J. Shiller's Yale Website. Additional data taken from US Census Bureau and US Department of Commerce.]