For those who are regulars on the Vanguard, these are the data that underlie a long time debate over the impact of the rate of growth in Davis on housing prices. Be forewarned, if you are allergic to math, you might want to skip to the graphs that illustrate some of the relationships between new building permits and housing prices. The data show in a number of different ways that building more in Davis does not reduce the price of housing in Davis when compared to Sacramento or other cities in the region.
The Mayor Sue Greenwald requested that someone analyze the relationship between the number of building permits and housing prices.
The price data come from RAND California through 2002 and since 2002, the data come from Trends in California Real Estate from the CA Realtors Association for July of each year.
The analysis looks at a number of different iterations here--current building permits, lagged building permits, all permits, or only single-family permits. None of these changes impact the overall findings.
The first chart here just shows the raw numbers.
There are some pretty interesting features of the raw numbers. You can see that housing permits peaked in the late 1990s with 1013 new permits in 1998 and 954 permits in 1999. Measure J comes in just after that and you can see the number of new permits really fall off after that point. You can also see the price of homes increases by a large margin after 1994. But as you can see, the price of homes in Sacramento increases as well over the same period of time. The last column is what is called the Davis premium--basically the ratio of housing prices in Davis divided by the housing prices in Sacramento.
Now that's the key variable right there, because that is used as an indicator for the regional housing prices.
As you can see in this graph, the two median sales prices of homes in Davis and Sacramento County track almost perfectly. Davis is higher throughout the period, but the increase in home prices is matched by the increase of home prices in Sacramento County. In fact, the correlation coefficient for the two is .99. The correlation coefficient for the non-statistics person is basically a measure of how interdependent two variables are to each other. The value indicates how much of a change in one variable is explained by a change in another. .99 means that 99% percent of the change in the Davis median sales price is explained by the change in the Sacramento County Median Sales price. That means it is almost a perfect relationship between the two. Frankly in my background in the social sciences, you are often happy if you can explain 30 or 40 percent of the relationship.
The next graph looks at the relationship between building permits and housing prices, once we control for the housing prices in other areas in the region. To do this, we use the Davis premium and plot it against the number of building permits. We use the lagged variable to approximate the time it takes for a home with an approved permit to actually hit the market. But as I said earlier, it does not matter if you use the lagged variable or unlagged variable.
Let's go back through this graph again. The Davis Premium is the median cost of housing in Davis divided by the median cost of housing in Sacramento County. The higher ratio means that Davis home prices are growing faster than Sacramento County home prices. The lower ration means that the Sacramento County home prices are growing faster than Davis home prices. Throughout most of the period the premium is between 1.4 and 1.6. That means that Davis home prices are 140% to 160% higher than Sacramento home prices. However, when we plot those ratios against the number of building permits we see that there is no relationship between the building permits and the relative price of housing in Davis. Even with two outliers, you have pretty much a flat ratio between the two.
Analysis and Implications
The argument on this blog has been that Davis is subject to the market forces of supply and demand just like any other community. That is correct, however, as some have contended the problem with such a simple view is that it does not take into account the much larger regional market. What we see in these data is that housing prices have gone up about the same rate in the fast growing market of Sacramento County as they have in the relatively slow growing market of Davis.
There is a differential between the cost of housing in Davis, but it is created largely by its desirability rather than its rate of growth.
It is important to note that these data do not resolve the issue of whether and how much Davis should grow. What it does suggest is that we cannot expect that more houses and more residential growth are going to result in a reduced cost of homes. Therefore, we should not tailor our growth policies in terms of number of units around the issue of affordability and the concern that middle class people are being priced out of Davis. We likely cannot change that by building more units unless we build so many units that we begin to reduce our quality of life.
If we built thousands of new homes, then we could possible begin to cut into the differential in housing prices between Davis and the surrounding communities. But how much would it take? And such an endeavor would likely result in a whole new set of problems. It is not as though much faster growing communities have fewer problems than we do in Davis. Nor is it true that they have better school systems than we do in Davis.
These data do have enormous implications for our land use policies however. They tell us that we are likely not going to gain affordability through growth. That means that if we are concerned about declining enrollment, we need to find ways to build the type of homes that young families want to live in. The natural starting point would be the university. The city of Davis clearly needs to work with the university in partnership to supply housing for young faculty members and staff--people who are likely to have children of school age. What we face in Davis is not unique to other college towns, we simply have to find ways to provide that kind of housing--not through faster growth, but through innovative approaches, that maintain the desirability of this town while providing real options for people who work at the university.
---Doug Paul Davis reporting
The Mayor Sue Greenwald requested that someone analyze the relationship between the number of building permits and housing prices.
The price data come from RAND California through 2002 and since 2002, the data come from Trends in California Real Estate from the CA Realtors Association for July of each year.
The analysis looks at a number of different iterations here--current building permits, lagged building permits, all permits, or only single-family permits. None of these changes impact the overall findings.
The first chart here just shows the raw numbers.
There are some pretty interesting features of the raw numbers. You can see that housing permits peaked in the late 1990s with 1013 new permits in 1998 and 954 permits in 1999. Measure J comes in just after that and you can see the number of new permits really fall off after that point. You can also see the price of homes increases by a large margin after 1994. But as you can see, the price of homes in Sacramento increases as well over the same period of time. The last column is what is called the Davis premium--basically the ratio of housing prices in Davis divided by the housing prices in Sacramento.
Now that's the key variable right there, because that is used as an indicator for the regional housing prices.
As you can see in this graph, the two median sales prices of homes in Davis and Sacramento County track almost perfectly. Davis is higher throughout the period, but the increase in home prices is matched by the increase of home prices in Sacramento County. In fact, the correlation coefficient for the two is .99. The correlation coefficient for the non-statistics person is basically a measure of how interdependent two variables are to each other. The value indicates how much of a change in one variable is explained by a change in another. .99 means that 99% percent of the change in the Davis median sales price is explained by the change in the Sacramento County Median Sales price. That means it is almost a perfect relationship between the two. Frankly in my background in the social sciences, you are often happy if you can explain 30 or 40 percent of the relationship.
The next graph looks at the relationship between building permits and housing prices, once we control for the housing prices in other areas in the region. To do this, we use the Davis premium and plot it against the number of building permits. We use the lagged variable to approximate the time it takes for a home with an approved permit to actually hit the market. But as I said earlier, it does not matter if you use the lagged variable or unlagged variable.
Let's go back through this graph again. The Davis Premium is the median cost of housing in Davis divided by the median cost of housing in Sacramento County. The higher ratio means that Davis home prices are growing faster than Sacramento County home prices. The lower ration means that the Sacramento County home prices are growing faster than Davis home prices. Throughout most of the period the premium is between 1.4 and 1.6. That means that Davis home prices are 140% to 160% higher than Sacramento home prices. However, when we plot those ratios against the number of building permits we see that there is no relationship between the building permits and the relative price of housing in Davis. Even with two outliers, you have pretty much a flat ratio between the two.
Analysis and Implications
The argument on this blog has been that Davis is subject to the market forces of supply and demand just like any other community. That is correct, however, as some have contended the problem with such a simple view is that it does not take into account the much larger regional market. What we see in these data is that housing prices have gone up about the same rate in the fast growing market of Sacramento County as they have in the relatively slow growing market of Davis.
There is a differential between the cost of housing in Davis, but it is created largely by its desirability rather than its rate of growth.
It is important to note that these data do not resolve the issue of whether and how much Davis should grow. What it does suggest is that we cannot expect that more houses and more residential growth are going to result in a reduced cost of homes. Therefore, we should not tailor our growth policies in terms of number of units around the issue of affordability and the concern that middle class people are being priced out of Davis. We likely cannot change that by building more units unless we build so many units that we begin to reduce our quality of life.
If we built thousands of new homes, then we could possible begin to cut into the differential in housing prices between Davis and the surrounding communities. But how much would it take? And such an endeavor would likely result in a whole new set of problems. It is not as though much faster growing communities have fewer problems than we do in Davis. Nor is it true that they have better school systems than we do in Davis.
These data do have enormous implications for our land use policies however. They tell us that we are likely not going to gain affordability through growth. That means that if we are concerned about declining enrollment, we need to find ways to build the type of homes that young families want to live in. The natural starting point would be the university. The city of Davis clearly needs to work with the university in partnership to supply housing for young faculty members and staff--people who are likely to have children of school age. What we face in Davis is not unique to other college towns, we simply have to find ways to provide that kind of housing--not through faster growth, but through innovative approaches, that maintain the desirability of this town while providing real options for people who work at the university.
---Doug Paul Davis reporting
