ANNIE E. CASEY FOUNDATION MAKING CONNECTIONS INITIATIVE HOUSING MARKETS IN CASEY METROS: WHAT HAS HAPPENED SINCE 2000? G. Thomas Kingsley and Beata Bajaj January 2005 THE URBAN INSTITUTE WASHINGTON, DC
HOUSING MARKETS IN CASEY METROS: WHAT HAS HAPPENED SINCE 2000? By G. Thomas Kingsley and Beata Bajaj The Urban Institute, January 2005 Introduction Expanding wealth creation through home ownership is a core objective in Making Connections. The ability to achieve it is influenced by conditions in the metropolitan housing market, as well as conditions in Making Connections neighborhoods themselves. In today s hot markets, metropolitan-wide pressures accelerate prices in low income neighborhoods as well more prosperous neighborhoods. It becomes more and more difficult even for working families to find housing they can occupy anywhere without devoting an exorbitant share of their income to paying for it. On the other hand, while an overheated market is to be avoided, some basic market strength remains essential to the goal. Where local housing prices are declining, homeownership can erode a family s equity rather than enhancing it. How do the metropolitan areas that are the focus on Annie E. Casey Foundation initiatives stack up along these lines in recent years? In this brief, we examine recent trends in one important market indicator: the changing relationship between metropolitan incomes and the sales prices of single family homes. 1 Data are presented first for 81 large metropolitan areas for which data are available, and then for the Casey metros, in the aggregate and individually (8 Making Connections sites and 4 current or former Civic Sites). 2 It would have been preferable if we could have examined changes in rent levels as well, since those impact lower-income families more directly, but rent data are not available. Nonetheless, the census shows that rent and home value changes are highly correlated; i.e., in a metropolitan area 1 Median incomes are as estimated annually by the U.S. Department of Housing and Urban Development (HUD). The estimating methodology, which relies on census data and information from other intercensal surveys, is documented at http://www.huduser.org\datasets\il.html. Data on median sales prices for single-family homes are from the National Association of Realtors (NAR) Existing Home Sales Survey (documented at: http://www.realtor.org/research.nsf/pages/ehsdata). The 81 metropolitan areas included are those among the 100 U.S. metros for which NAR survey data are available. 2 The Casey metros examined in this paper include eight Making Connections sites (Denver, Hartford, Indianapolis, Louisville, Milwaukee, Oakland, Providence, San Antonio, and Seattle data are not available for Des Moines and Oakland) and four current or former Civic Sites (Atlanta, Baltimore, New Haven, and Washington DC).
- 2 - where home prices are outstripping incomes, rents are generally likely to be accelerating rapidly as well. The National Story In the late 1990s, almost all economic indicators, including those pertaining to housing, were looking up. For the 81 large metropolitan areas on average between 1995 and 2000, median family incomes and the median sales price of single family homes both increased by a healthy rate, just over 2 percent per year (Table 1). In 2000, the median sales price ($153,100) was 2.6 times the median family income ($58,500). 3 The economy (and family incomes), however, began to slow down after that. Over the next three years, income growth dropped considerably to an annual rate of 1.0 percent, less than half of what it had been in the earlier period. But what was unusual was that housing prices did not slow down as well. In fact, the median sales price accelerated, growing at an annual rate of 5.2 percent from 2000 to 2003, more than double the rate of the late 1990s. Traditionally, the housing sector almost always performs badly when the national economy does. In fact, a slowing of housing price increases has been considered a harbinger of trouble for the broader economy. What happened this time? Most observers credit low interest rates as the most important factor in maintaining dynamism in housing when the economy became sluggish, enough so that housing actually cushioned the overall economic downturn. 4 Regardless of the mechanisms at work, the results were a serious blow to housing affordability. By 2003, the median sales price had gone up to $183,200 while median income only went up to $60,300. Whereas in 2000 it had taken 2.6 times the median income to purchase a home at the median price, by 2003 it took 3.0 times the median income to purchase the median home. 3 All income and price data reported in this brief are in constant 2003 dollars. 4 See, for example, the State of the Nation s Housing, 2004, by the Joint Center for Housing Studies (Cambridge, MA: Harvard University) which states that wealth effects related to home price appreciation, realized capital gains, and heavy home-equity borrowing appear to have contributed more than one quarter to the growth in personal consumption in both 2002 and 2003.
- 3 - Table 1 TRENDS IN MEDIAN SALES PRICE OF SINGLE FAMILY HOMES AND MEDIAN FAMILY INCOME Casey Metros and 81 Large Metros Values Median Median price Change median sales income Change median Median sales price/ (thous.) price (%/yr.) (thous.) income (%/yr.) median income 2003 1995-00 2000-03 2003 1995-00 2000-03 2000 2003 81 Large Metros 183.2 2.1 5.2 60.3 2.2 1.0 2.6 3.0 Casey metros Total 197.4 2.4 6.7 66.3 2.3 1.1 2.5 2.9 Making Connections 187.0 2.6 5.4 62.9 2.3 1.4 2.6 2.9 Civic sites 218.2 2.0 9.2 73.0 2.3 0.6 2.3 3.0 Making Connections Denver 238.2 6.5 4.2 68.0 2.5 0.8 3.2 3.5 Hartford 202.3 1.2 5.8 73.0 0.6 3.7 2.6 2.8 Indianapolis 121.1 1.0 0.3 62.9 2.8 0.7 1.9 1.9 Louisville 131.7 3.6 1.8 56.2 3.8 0.7 2.3 2.3 Milwaukee 182.1 1.6 6.6 62.6 3.2-1.5 2.3 2.9 Providence 233.4 1.1 16.6 58.4 1.2 3.2 2.8 4.0 San Antonio 118.1 1.0 4.8 50.5 1.8 3.1 2.2 2.3 Seattle 268.8 5.1 3.0 71.9 2.5 0.8 3.5 3.7 Civic Sites Atlanta 152.4 3.5 2.8 68.8 2.8 0.7 2.1 2.2 Baltimore 208.9 4.0 8.5 67.3 2.5-0.1 2.4 3.1 New Haven 225.3-0.1 11.6 71.0 0.8 3.1 2.5 3.2 Washington 286.2 0.6 13.6 84.8 3.2-1.4 2.2 3.4 Rank among the 81 large metros Making Connections Denver 16 4 36 16 32 40 12 18 Hartford 23 54 29 7 76 9 25 30 Indianapolis 65 59 76 30 21 44 71 76 Louisville 55 22 60 51 8 42 45 54 Milwaukee 26 45 26 31 15 71 42 28 Providence 17 56 4 44 67 13 17 14 San Antonio 68 58 33 72 53 14 49 56 Seattle 11 8 45 8 33 41 9 16 Civic Sites Atlanta 38 24 47 13 26 43 63 59 Baltimore 22 19 17 18 34 61 36 24 New Haven 20 69 13 9 73 15 31 23 Washington 10 63 9 2 16 70 55 20 Sources and definitions: see text
- 4 - Figure 1. Ratio of Median Sales Prices of Single-Family Homes to Median Incomes, 2000 and 2003 Average 81 metros Seattle 2000 2003 Denver Providence Hartford New Haven Baltimore Milwaukee Louisville San Antonio Washington, DC Atlanta Indianapolis 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Source: See text Variations Across Casey Metros Averages for the Casey metros together on all measures noted above look similar to the averages for all 81 large metros. The story is the same: solid growth in both incomes and housing prices leading up to 2000, then a notable slowing down of income growth while housing prices experienced an even more rapid ascent. This basic story is applicable for most individual Casey metros, but there are some marked variations within it. Across the Casey metros in 2003, median incomes range from a low of $50,500 (San Antonio, ranking 72nd among the 81 large metro areas) all the way up to $84,800 (Washington, 2nd highest) Table 1. Median sales prices range from a low of $118,100 (San Antonio, ranking 68 th ) up to a high of $286,200 (Washington, ranking 10 th ). These absolute levels,
- 5 - however, do not say anything about comparative well-being. An area with quite high housing costs could be in a good position if incomes were high enough and growing rapidly. Accordingly, it is important to focus on income growth and the relationship between incomes and housing prices in each metro. As to income growth, the experience of four of these metros (Hartford, Providence, San Antonio, and New Haven) did not conform to the basic story. They had comparatively modest income growth in the late 1990s, and their rates of income growth actually increased some over the 2000-2003 period. All other metros had experienced more rapid income growth in the earlier period and a sharp drop off after 2000. To avoid concentrating too much on the fluctuations, it is worth looking at the comparative growth over the entire 1995-2003 period. Here we find that some of the metros with the lowest absolute levels of income were experiencing the most rapid income growth. The leader among Casey metros was Louisville with annual real growth in median income averaging 2.6 percent. Others with rates above 2.0 percent included San Antonio, Indianapolis, and Atlanta. Those with the slowest income growth were Baltimore, Washington, and Milwaukee (rates in the 1.4 to 1.5 percent range). As to the affordability of single-family homes, the range is also wide Figure 1. In 2003, the least affordable Casey metro by these measures was Providence where the median sales price was 4.0 times the median income. Next in order were Seattle (3.7), Denver (3.5), Washington (3.4), New Haven (3.2) and Baltimore (3.1). All of these fell in the top third of the 81 metros, but none of them was among the very highest nationally. Several other metros had ratios in excess of 5.0 (including New York, Miami, and several California metros including San Francisco and Los Angeles). While we do not have separate data for Oakland, it is very likely that since it is a part of the greater San Francisco region, it is the least affordable Casey metro. At the other end of the spectrum, the most affordable Casey Metro was Indianapolis, where the median sales price was less than twice (1.9 times) median income (76 th out of 81). The next three most affordable were Atlanta, Louisville, and San Antonio (range of 2.2 to 2.4). (The three weakest markets nationally, with ratios below 1.8, were Syracuse, Rochester and Buffalo). The trends in these ratios also exhibited substantial variation. In general, the metros that were the least affordable to begin with have suffered the most notable increases in pressure; i.e., affordability ratios have increased the most. Most disturbing was Washington where the 2003 ratio was 53 percent higher than what it had been in 2000; that for Providence was 44 percent higher. Baltimore, New Haven and Milwaukee also saw notable increases (27-28 percent higher).
- 6 - At the other extreme, several metros that had the least serious affordability problems to begin with experienced the most favorable trends. The affordability ratio for Indianapolis actually decreased slightly from 2000 to 2003. For three others in this group, ratios only increased modestly (less than 10 percent): Louisville, San Antonio, and Hartford. Seattle represents a special case. Compared to other Casey metros, it probably suffered most from the economic downturn of the early part of this decade and this problem did spill over into the housing sector. While it was still the second least affordable Casey metro (housing price 3.7 times income in 2003) that ratio had not increased much from what it had been in 2000. In rounding out this analysis it is important to note again that we have only measured affordability in relation to the sales prices of single-family homes. While, as we noted, changes in sales prices and rents are generally correlated, differences can be sizeable in some cases. In particularly, this raises cautions about the interpretation of our numbers for Hartford, New Haven and Providence where renters make up a higher share of all households than is typically the case. The data show that these three markets are tough for would-be home buyers, but we cannot be sure about the extent to which that is the case for renters. Expectations and Implications Few housing analysts think that the recent run-ups of housing prices in U.S. metros represent bubbles that are likely to burst and cause abrupt declines. However, most expect some moderation in the growth of home prices due to expected increases in interest rates nationally. 5 This is not a reason for complacency, however. Price-to-income ratios have grown to extreme levels in many urban areas and policy responses to moderate them are going to be essential in those areas. Among the Casey metros, this issue deserves attention where price/income ratios remain high by national standards, and acceleration has not abated: certainly in Oakland, Providence, Denver, Washington, New Haven and Baltimore. What needs to be done? To start, it is important to recognize that real solutions require region-wide responses and in most cases this has to include an increase in the aggregate rate of housing production. Subsidy funds are not going to be adequate to bridge the gap anywhere. 5 See the discussion of this issue as well in the State of the Nation s Housing, 2004, by the Joint Center for Housing Studies (Cambridge, MA: Harvard University). An abrupt drop in house prices due to a breaking bubble is likely only when there is a major decline in a region s economy, and few see this on the horizon for any U.S. metros at this point.
- 7 - If prices are to moderate, supply will have to expand faster than demand. A recent examination of the issue for metropolitan Washington recommends an approach with three main elements. 6 Expanding housing production in higher-density, mixed-income developments in the central city and inner suburbs, to accommodate growth without encouraging sprawl. Preserving and expanding affordable housing choices throughout the region to enable lower-income families to live closer to their jobs. Promoting more balanced job growth to ease pressures on the highest-cost suburban communities and expand employment opportunities in areas where housing affordability options are more abundant. As regional solutions are being worked out, however, individual central cities with hot housing markets should be taking steps to diminish pressures for their lowest income residents. The main idea is not to prevent new investment, but to manage it so that it benefits original residents rather than displacing them. Possible actions range from the implementation of inclusionary zoning (requiring that a share of the units in new development projects be offered at rents affordable to low-income families) to serious monitoring of HUD-assisted resources and developing strategies to prevent loss. 7 What about the weak market metros (where property values may actually be declining in the inner-city)? More analysis is needed to find out if any Making Connections are actually in this state, 8 but the data presented here suggest it may be a possibility at least for troubled neighborhoods in Indianapolis, Louisville and San Antonio. Ultimately, the solutions for these markets depend on regional actions to reinvigorate the metropolitan economy. Whatever the success on that score, however, almost all cities have seen some reinvestment in high-poverty neighborhoods of late. Community development practitioners today stress partnership approaches (involving businesses and large anchor institutions) and being very realistic regarding adjusting to local market conditions. 9 6 Housing in the Nation s Capitol, 2004, by Margery A. Turner, G. Thomas Kingsley, Kathryn L.S. Pettit, and Noah Sawyer. (Washington, DC: The Fannie Mae Foundation, 2004). 7 For a broader menu of policy and program options, see Rethinking Local Affordable Housing Strategies: Lessons from 70 Years of Policy and Practice, by Bruce Katz and Margery A. Turner (Washington, DC: The Brookings Institution, 2003). 8 The Urban Institute has begun updating its Home Mortgage Disclosure Act (HMDA) data files and plans to use these soon to address this question for Making Connections neighborhoods. 9 See Building a New Framework for Community Development in Weak Market Cities, by Paul C. Brophy and Kim Burnett (Denver, CO: The Community Development Partnership Network, April 2003).