In my last column, I sent out an all-points bulletin requesting any information on a data series that measures regulatory costs by county in California. I even promised four free tickets to our 40th Annual Economic Forecast Conference on Dec. 6 to anyone who could supply the data. Even better, I promised a shoutout at the conference to the person who delivered the goods.
Despite that incredibly generous package of perks, I had no takers. But totally unrelated to my search for a regulatory cost index, I was trying to solve an econometric problem and recalled that a Wall Street Journal article I’d read several years ago dealt with it. Imagine my surprise when I discovered that in addition to explaining my econometric problem, the article also made reference to a data series on regulatory costs.
The regulatory cost series in question is the Wharton Residential Land Use Regulatory Index, or WRLURI. The index uses factor analysis to capture the degree of stringency of local regulatory environments. Like most data series, it’s not perfect. It’s a bit dated, since the series hasn’t been revised since 2008. And although it covers 2,600 communities, the number of areas sampled in some counties is too low for reliable statistical tests. But it’s something.
To remind you about the significance of this data, recall that in a series of three of my On California columns, I’ve tried to show how the wide range of housing prices among California’s counties can be explained by a few variables. I found, for example, that income and natural amenities explain about 81% of the variation in home prices. That’s pretty good, but… not good enough. My equation, for example, projected an Orange County home price of $662,334—about $40,000 short of the actual median of $702,918.
That’s when I came up with the bright idea that differences in regulatory costs across California counties might explain the shortfall. But I needed a regulatory cost series to add to my explanatory equation. That’s why my serendipitous discovery of WRLURI was so very exciting. I get chills just recounting the experience.
Sadly, the test results didn’t support my theory. While the regression indicates that greater regulation increased home prices, the impact was small and not statistically significant. Even the proportion of variation in home prices explained by the equation with two variables (income and amenities), which stood at 81%, did not improve with the addition of the third variable (regulatory costs).
How can that be? A scatter diagram showing only the relationship between home prices and regulatory costs clearly indicated a positive relationship. For some reason that relationship wasn’t being picked up in the equation. Then it hit me. Regulatory costs were moving pretty much in tandem with the series that measures amenities (see graph). Because of that, these two variables were what we econometricians call ‘collinear,’ and the resulting equation put all of the explanatory power in the amenities rather than the regulatory series. In other words, the natural amenities series already served as a proxy for land use regulations. As a result, there was no need for a separate regulatory measure in the equation.
While this means that I haven’t gotten my equation any closer to my goal of explaining OC’s median home price, I have uncovered a totally new line of research: namely, that counties with greater amenities tend to impose more regulations. This suggests that NIMBY-type attitudes that push local governments to impose more stringent land use restrictions occur more often in areas with more attractive amenities, like nicer weather, cleaner air and beautiful surroundings. As a result, there will be more regulations in a locale like OC than let’s say … Tulare County (please don’t pass this article to any residents in said county).
Now that may not be the stuff of a Nobel Prize, but it has all the makings of something that will keep me busy for a while. The neat thing about research—it doesn’t move in a straight line. There are all kinds of strange twists and turns that pull one hither and yon, kind of like a sailboat headed for a distant point but being buffeted by winds along the way.
But let’s get back to that distant point, namely explaining OC’s median housing price. While I’m 81% of the way there, I want to get closer, and guess what? I have another idea, and I think it’s even better than my failed regulatory costs theory. Regrettably, I can’t report on it now, since I’m in the midst of testing that theory. I should have something for you in my next On California.
I don’t know about you, but I’m having a blast!
