Real Estate Data for the Rest of Us

Housing Barometer: Recovery Continues, But Virtuous Cycle Not So Saintly

Jed Kolko, Chief Economist
October 1, 2014

Three out of five Housing Barometer measures are getting close to normal. But the two measures that hitch housing to the broader economy are still struggling, so the job market and housing market aren’t helping each other as they should.

How We Track This Uneven Recovery

Since February 2012, Trulia’s Housing Barometer has charted how quickly the housing market is returning to “normal” based on multiple indicators. Because the recovery is uneven, with some housing activities improving faster than others, our Barometer highlights five measures:

  1. Existing home sales, excluding distressed sales (National Association of Realtors, NAR)
  2. Home-price levels relative to fundamentals (Trulia Bubble Watch)
  3. Delinquency + foreclosure rate (Black Knight, formerly LPS)
  4. New construction starts (Census)
  5. The employment rate for 25-34 year-olds, a key age group for household formation and first-time homeownership (Bureau of Labor Statistics, BLS)

Home prices from our Bubble Watch is a quarterly report. The other four measures are reported monthly. To reduce volatility, we use three-month moving averages for these measures. For each indicator, we compare the latest available data to (1) its worst reading during the housing bust and (2) its pre-bubble “normal” level.


All Five Measures Improved Year-Over-Year

Four of the five Housing Barometer indicators made good progress over the past year and the fifth – non-distressed existing home sales – eked out a slight increase. But, despite improvement, the employment rate for young adults still hasn’t gotten even half of the way back to normal.

Housing Indicators: How Far Back to Normal?

Now One quarter ago One year ago
Existing home sales, excl. distressed 80% 64% 79%
Home price level 75% 66% 56%
Delinquency + foreclosure rate 74% 74% 56%
New construction starts 49% 49% 37%
Employment rate, 25-34 year-olds 37% 35% 25%
For each indicator, we compare the latest available data to (1) its worst reading for that indicator during the housing bust and (2) its pre-bubble “normal” level
  • Existing home sales (excluding distressed) were 80% back to normal in August, up from 64% one quarter earlier, after stumbling from 79% one year ago. Distressed sales keep falling. Increasingly, foreclosures are concentrated in states with a foreclosure laws that result in a longer legal process. Existing sales are in better shape than new home sales, dominating the market even more than usual. The ratio of existing to new home sales was 10:1 in August, compared with a long-term normal ratio of 6:1..
  • Home prices continue to climb, though at a slower rate. Trulia’s Bubble Watch shows prices were 3.4% undervalued in 2014 Q3, compared with 13.5% undervalued at the worst of the housing bust. That means prices are three-fourths of the way back to their “normal” level at which they’re neither over- nor undervalued.
  • The delinquency + foreclosure rate was 74% back to normal in August, the same as one quarter ago and up significantly from 56% one year ago. With the share of mortgage borrowers with negative or near-negative equity dropping, the default rate should continue to go down.
  • New construction starts are 49% back to normal, the same as one quarter ago and up from 37% one year ago. Multi-unit starts continue to lead the construction recovery. Year-to-date multi-unit starts are up 23% year-over-year, versus just 3% for single-family starts. Even though single-family starts are far below normal levels, household formation looks too weak to support more single-family homebuilding.
  • Employment for young adults brings up the rear. August’s three-month moving average shows that 75.7% of adults age 25-34 are employed, which is just 37% of the way back to normal. Because young adults need jobs in order to move out of their parents’ homes, form their own households, and eventually become homeowners, the housing recovery depends on millennials finding work. Among 25-34 year-olds, just 12% who have jobs live with their parents. By contrast, 21% without jobs do.

The Housing Market and the Broader Economy Aren’t Helping Each Other

The two lagging Housing Barometer measures – construction and young-adult employment – connect the housing market to the job market. First, housing should help jobs: construction adds to employment not only in homebuilding but also in related industries like furniture manufacturing and home-improvement retailing. Second, jobs should help housing: young adults are more likely to rent or buy, rather than live with others, if they have jobs. In this recovery, young-adult employment and construction are weak – so the virtuous cycle of housing and jobs isn’t looking quite so virtuous.

That’s not to say that housing isn’t doing anything for the economy. Rising home prices make homeowners wealthier, and the more wealth people have, the more they spend. And the decline in defaults and foreclosures have helped stabilize the financial system and hard-hit neighborhoods. As we’ve seen, home prices right themselves, as undervalued homes attract investors and other buyers, pushing prices back up. In turn, higher prices make defaults less likely.

But as the housing recovery continues, it depends less on the “rebound effect” – this tendency of the housing prices to right themselves – and more on such fundamentals as jobs, income growth, and household formation. These have been slow to improve in this recovery. In particular, the Housing Barometer shows that young-adult employment lags. What’s more, new Census data showed that median income has stagnated and household formation is far below normal levels. In this recovery, jobs and housing can’t get what they need from each other.


NOTE: Trulia’s Housing Barometer tracks five measures: existing home sales excluding distressed (NAR), home prices (Trulia Bubble Watch), delinquency + foreclosure rate (Black Knight), new home starts (Census), and the employment rate for 25-34 year-olds (BLS). Also, our estimate of the “normal” share of sales that are distressed is 5%; Black Knight reports that the share was in the 3-5% range during the bubble. For each measure, we compare the latest available data to (1) the worst reading for that indicator during the housing bust and (2) its pre-bubble “normal” level. We use a three-month average to smooth volatility for the four indicators that are reported monthly (all but home prices). The latest data are from August for the employment rate, existing home sales, new construction starts, and the delinquency + foreclosure rate; and Q3 for home prices.


Bubble Watch: Home Prices 3% Undervalued, With Few Metros Bubbling Up

Jed Kolko, Chief Economist
October 1, 2014

Home prices now look 3% undervalued measured by long-term fundamentals. Just 7 of the 100 largest metros are more than 10% overvalued.

Trulia’s Bubble Watch shows whether home prices are overvalued or undervalued relative to their fundamental value by comparing prices today with historical prices, incomes, and rents. The more prices are overvalued relative to fundamentals, the closer we are to a housing bubble – and the bigger the risk of a price crash. Sharply rising prices aren’t necessarily a sign of a bubble. By definition, a bubble develops when prices look high relative to fundamentals.

Bubble watching is as much an art as a science because there’s no definitive measure of fundamental value. To try to put numbers on it, we look at the price-to-income ratio, the price-to-rent ratio, and prices relative to their long-term trends. We use multiple data sources, including the Trulia Price Monitor, as leading indicators of where home prices are heading. We combine these various measures of fundamental value rather than relying on a single factor because no one measure is perfect. Trulia’s first Bubble Watch report, from May 2013, explains our methodology in detail. Here’s what we found this quarter. (This report contains larger-than-usual revisions of previous Bubble Watch estimates. See note.)

Home Prices are 3% Undervalued Nationally

We estimate that home prices nationally are 3% undervalued in the third quarter of 2014 (2014 Q3). In 2006 Q1, during the past decade’s housing bubble, home prices soared to 34% overvalued before dropping to 13% undervalued in 2012 Q1. One quarter ago (2014 Q2), prices looked 5% undervalued; one year ago (2013 Q3), prices looked 6% undervalued. This chart shows how far current prices are from a bubble:


Texas and California Metros Look Most Overvalued

The most overvalued market is now Austin, at 19%, followed by the California metros of Los Angeles, Orange County, San Francisco, and Riverside-San Bernardino. The California metros on the top-10 list were all significantly overvalued during the past bubble, ranging from 46% overvalued in San Francisco to a dizzying 87% in Riverside-San Bernardino. By contrast, Austin and Houston are the only metros out of the 100 largest that look more overvalued today than in 2006. Texas markets avoided the worst of the housing bubble during the past decade. Recently, they’ve had double-digit home-price increases.

Top 10 Metros Where Home Prices Are Most Overvalued
# U.S. Metro Home prices relative to fundamentals, 2014 Q3 Home prices relative to fundamentals, 2006 Q1 Year-over-year change in asking prices, Aug 2014
1 Austin, TX +19% +2% 11.9%
2 Los Angeles, CA +15% +73% 8.9%
3 Orange County, CA +15% +66% 6.0%
4 San Francisco, CA +12% +46% 11.2%
5 Riverside-San Bernardino, CA +11% +87% 13.8%
6 Honolulu, HI +10% +36% 6.7%
7 San Jose, CA +10% +53% 10.4%
8 Houston, TX +8% +1% 10.4%
9 Denver, CO +7% +17% 9.4%
10 Oakland, CA +7% +67% 12.4%
Note: positive numbers indicate overvalued prices; negative numbers indicate undervalued, among the 100 largest metros. Click here to see the price valuation for all 100 metros: Excel or PDF.

Almost all of the most undervalued metros today are in the Midwest and New England, led by Dayton and Cleveland. One year ago, Las Vegas and two Florida metros, Lakeland-Winter Haven and Palm Bay-Melbourne-Titusville, were on the most-undervalued list. Since then, price gains have lifted them off this list. In the past year, price gains in the undervalued Midwestern markets like Detroit have outpaced price gains in the undervalued New England markets like New Haven.

Top 10 Metros Where Home Prices Are Most Undervalued
# U.S. Metro Home prices relative to fundamentals, 2014 Q2 Home prices relative to fundamentals, 2006 Q1 Year-over-year change in asking prices, May 2014
1 Dayton, OH -21% 8% 8.8%
2 Cleveland, OH -19% 13% 5.9%
3 Detroit, MI -18% 33% 11.9%
4 Akron, OH -18% 13% 8.9%
5 Lake County-Kenosha County, IL-WI -17% 24% 12.2%
6 Toledo, OH -17% 17% 9.6%
7 New Haven, CT -16% 31% -0.9%
8 Camden, NJ -15% 31% 1.8%
9 Worcester, MA -15% 38% 3.5%
10 Fairfield County, CT -14% 30% 0.4%

Note: positive numbers indicate overvalued prices; negative numbers indicate undervalued, among the 100 largest metros. Click here to see the price valuation for all 100 metros: Excel or PDF.

Are We Headed Toward The Next Bubble?

One test of whether it’s time to sound the bubble alarm is whether prices are rising faster in markets that are already overvalued. Price gains in overvalued markets are a sign that we’re headed for danger, while price gains in undervalued markets are probably just a sign of getting back toward normal.

To measure this, we compare the most recent year-over-year asking-price change from the Trulia Price Monitor with our Bubble Watch measure from 2013 Q3, one year ago. That’s because what matters is whether overvalued markets subsequently see faster price gains (remember that current Bubble Watch values, by design, incorporate recent price trends).

The scatterplot below shows the relationship. Hard to see a pattern, right? Actually, there’s a negative relationship, but it’s small (correlation = -0.07) and not statistically significant. At least we can say that overvalued markets are not systematically seeing larger price increases, though some individual overvalued markets like Austin and Riverside-San Bernardino did have big price jumps.


Another measure of bubble risk is how many markets are more than 10% overvalued. As of 2014 Q3, only seven of the top 100 metros exceeded this level, as shown in the table above. That’s the highest number since 2009 Q1, when prices were plummeting and the past bubble had mostly deflated. The last time that the number of 10%+ overvalued markets was at least seven and rising was 2000 Q2 – early in the formation of that bubble.


All this means that bubbles should not be our top housing worry today. Our latest Housing Barometer shows that weak construction and subpar young-adult employment are the recovery’s big red flags. By contrast, prices are slowing to a sustainable pace and staying within striking distance of normal.


Note: each quarter’s Bubble Watch includes revisions to previous estimates because the underlying data are often revised or updated. To compare the national or metro trend over time, look at the current report’s historical numbers, not previously reported numbers. This quarter’s Bubble Watch contains larger-than-usual revisions because a key input data series – the Case-Shiller national index – recently had significant revisions that resulted in less extreme price swings during the boom and bust.


Are We Building Too Many Single-Family Homes?

The vacancy rate for single-family homes increased in 2013 and remains well above bubble and pre-bubble levels.

Jed Kolko, Chief Economist
September 17, 2014

What? Too much new single-family construction? It sounds hard to believe, with only 618,000 single-family housing starts in 2013, heading toward 622,000 in 2014 – far below the pre-bubble average of 1.1 million per year in the 1990s. Even when adding in multi-unit building, which is booming, construction remains a laggard in the housing recovery and is contributing less than it should to employment and economic growth.

Of course, the historical norm doesn’t tell us what the just-right level of construction is now. That depends on the rate at which new households are formed. If new construction runs ahead of household formation, more homes sit empty and the vacancy rate rises. In 2004 and 2005, during the bubble, construction of single-family homes soared to over 1.5 million units. Then, during the bust, household formation slowed, in part because more young people lived with parents. Too much housing and too few households were a dangerous cocktail during the housing bust and recession, causing the vacancy rate to climb until 2010. Since then, the vacancy rate has fallen, but single-family construction has continued to wallow near all-time lows.

Newly released data from the Census Bureau’s American Community Survey (ACS) show that the vacancy rate for single-family homes actually ticked up a bit in 2013. That’s a big surprise. It suggests even today’s low level of single-family construction might still be too much, too soon. To determine whether we’re building too many homes, we need first to understand household formation, and then the vacancy rate.

Single-Family Rentals Increased Despite Low Household Formation Rate
To understand what’s happening with vacancy rates, let’s start by looking at changes in households and housing units in the past year broken down by owner-occupied and rented, and single-family and multi-unit:

Type of unit Change, 2012 to 2013, ‘000s Change, 2012 to 2013, % Change, 2006 to 2013, ‘000s Change, 2006 to 2013, %
Owner-occupied single-family -184 -0.3% -428 -0.7%
Renter-occupied single-family 331 2.3% 3540 31.2%
Owner-occupied multi-unit (i.e. condos) 18 0.5% -269 -6.4%
Renter-occupied multi-unit (i.e. apartments) 263 1.0% 2259 9.7%
Total single-family units, incl. vacant 226 0.3% 4701 5.5%
Total multi-family units, incl. vacant 199 0.6% 2131 6.5%
Total housing units, incl. vacant 356 0.3% 6496 5.1%
Total households 321 0.3% 4674 4.2%
Note: total housing units and total households include mobile homes, boats, RV’s, vans, etc. and their occupants.

… continue reading


American Mobility Remains Stuck in Low Gear

11.7% of Americans moved in the past year, unchanged from the previous year. But more people moved in search of cheaper housing.

Jed Kolko, Chief Economist
September 17, 2014

Yesterday, the Census released the Current Population Survey (CPS) data, giving an up-to-date picture on how many Americans are moving, how far they’re going, and why they’re making that move. (See note.) The mobility rate remains at a low level: 11.7% of Americans moved in the year ending March 2014, unchanged from the year ending March 2013.

At this mobility rate, the typical American stays put eight and a half years between moves. Remember the old rule of thumb that people move every seven years? Well, that was true until around 2003. In fact, the mobility rate has been falling for decades, as we pointed out in this post last year. Back in the 1950s and 1960s, Americans moved every five years on average. That rose to every seven years by the turn of the century and has since increased to the current eight-and-a- half year rate.

In today’s post, we look at the 2014 data to highlight the most recent mobility trends.

No Reversal in the Long-Term Mobility Decline
With the percentage of Americans moving stuck at 11.7% in 2014, mobility remains near the all-time low of 11.6% in 2011. That’s considerably below the 14% rate from the early 2000s. The housing bust and recession offer possible explanations why people are stuck in place – things like negative home equity and few job opportunities to move for. Still, mobility also declined both before and during the housing bubble. Furthermore, mobility has barely budged since 2011 despite a significant drop in the percentage of borrowers with negative equity and a modest recovery in the job market.

MobilityRate … continue reading


More Millennials Leave Parental Nest, Without Lifting Housing Market

Although slightly fewer young adults are living in their parents’ homes, don’t get too excited. Fewer are heading their own households, and the true young adult homeownership rate slipped in 2014.

Jed Kolko, Chief Economist
September 16, 2014

This morning, the Census Bureau released 2014 data that show whether Americans own, rent, or live under someone else’s roof. (See note.) As we’ve pointed out before, the published homeownership rate is often a misleading guide to what’s really happening in the housing market. For instance, suppose young people move out of their parents’ homes into rental apartments. That would lower the published homeownership rate because the number of renters has increased – even though the number of young homeowners is unchanged.

Using these fresh 2014 data, we update several key measures of housing and living arrangements, including:

  1. the percentage of young people living with their parents;
  2. the headship rate, which is the percentage of adults who head a household, either as an owner or a renter;
  3. our “true” homeownership rate, which equals the percentage of adults who are homeowners.

These three measures are closely related. If young people move out of their parents’ homes and become either renters or homeowners, the share of young adults living with parents goes down, while the headship rate for young adults goes up. Furthermore, the true homeownership rate equals the published homeownership rate times the headship rate – and therefore takes into account whether people are dropping out of or entering the housing market. Thus, it gives a clearer picture of whether the housing market is recovering. With that overview, here’s what the new 2014 data show.

True Young-Adult Homeownership Rate Falls in 2014, Reversing 2013 Increase
Let’s start with those millennials in the basement. They’re still there, but, ever so slowly, more are moving out. In 2014, 31.1% of 18-34 year-olds lived with their parents, down slightly from 31.2% in 2013 and from the peak of 31.6% in 2012.

1834LivingWithParents … continue reading

1 comment