In Depth Analysis: CalculatedRisk Newsletter on Real Estate (Ad Free) Read it here.

Sunday, November 18, 2007

The Fed, Household Real Estate Assets and Equity

by Calculated Risk on 11/18/2007 12:37:00 PM

JW sent me this Washington Post commentary: Bits of Bad News Obscure A Big Truth About Wealth

Despite declining prices in many markets, homeowners still control near-record equity holdings, just under $11 trillion.

... there's no question that equity holdings have declined recently and may well be lower when the Fed issues its next quarterly report, in mid-December. But in an $11 trillion marketplace, a $6 billion giveback in a cyclical correction is not a cause for panic.
This is similar to the argument that John Berry of Bloomberg suggested a few days ago: Bloomberg's Berry: No Recession. Although the WaPo's Kenneth Harney didn't focus on equity extraction - just wealth - it is declining wealth and less equity extraction that will impact the U.S. economy in the coming quarters. After reading the Bloomberg article, I tried to show, using the Fed's numbers that the Home ATM is running out of cash.

But Harney raises another interesting question: Why does the Fed Flow of Funds report show a decline in household real estate equity of only $6 Billion in Q2? Here is the relevant table (B.100 Balance Sheet of Households and Nonprofit Organizations).

The Fed shows household real estate assets (line 4) increased from $20.808 Trillion in Q1 to $20.997 Trillion in Q2. An increase of $189 Billion in assets. Wait - aren't prices falling? I'll get back to that ...

The Fed also shows household mortgages (line 32) increased from 9.951 Trillion in Q1 to $10.146 Trillion in Q2. An increase in mortgages of $195 Billion. The Home ATM was very active in Q2!

So that is the $6 Billion decline in equity mentioned by Harney: An increase in assets of $189 Billion eclipsed slightly by an increase in mortgage debt of $195 Billion.

Why did assets increase $189 Billion in Q2? All the data isn't publicly available, and the calculation is very complicated, but here is a simple formula:

Assets (Q2) = Assets (Q1) * PriceChange (Q2) + Assets Added (New Homes, Home improvement) - Assets Subtracted (demolished, damaged, etc.)
I believe the key is the price change in Q2. Look at these price indices:

Household Distribution by Valuation Click on graph for larger image.

These are all nominal prices, and the OFHEO index is Purchase Only, seasonally adjusted, and set to 100 in Q1 2000. Note: The Case-Shiller composite indices are monthly, and the graph is plotted quarterly.

All of the Case-Shiller indices show house prices have peaked, and have been declining for over a year. Meanwhile the OFHEO purchase only seasonally adjusted index shows prices increased in Q2 by 0.5% (a 2.1% annual rate).

There are significant differences between the OFHEO HPI and the Case-Shiller National index. This is an excellent summary by OFHEO economist Andrew Leventis: A Note on the Differences between the OFHEO and S&P/Case-Shiller House Price Indexes
OFHEO’s House Price Indexes (the “HPI”) and home price indexes produced by S&P/Case-Shiller are constructed using the same basic methodology. Both use the repeat-valuations framework initially proposed in the 1960s and later enhanced by Karl Case and Robert Shiller. Important differences between the indexes remain, however. The two models use different data sources and implement the mechanics of the basic algorithm in distinct ways.
Based on a review of the methodology documentation that is available, it appears that OFHEO’s national index has broader geographic coverage than the S&P/Case-Shiller National Home Price Index. According to the methodology materials, the S&P/Case-Shiller index does not include house price data from thirteen states.
The S&P/Case-Shiller index also apparently has incomplete coverage in another 29 states. ... To the extent that the missing areas tend to be more rural counties, given that rural areas appear to be exhibiting stronger market conditions in recent periods, the missing data might partially explain why the OFHEO and S&P/Case-Shiller national indexes diverge.
OFHEO’s sales price data include only homes that have conforming mortgages, while the S&P/Case-Shiller sales prices are not restricted to houses with certain types of financing. Because many of the homes not covered in the OFHEO index may be relatively expensive (i.e., may have required non-conforming, “jumbo” mortgages), the OFHEO restriction to conforming mortgages may produce appreciation rate estimates that are less reflective of price trends for the most expensive homes.
The OFHEO note suggests that the primary reason for the difference between Case-Shiller and OFHEO price indices is geographical coverage (not the loan limitations for OFHEO).

This is important because it appears the Fed uses the OFHEO index (or something similar) to calculate changes in household real estate asset value. If the Case-Shiller index is more representative of recent price changes, then the Fed actually underestimated the recent increase in household real estate assets!

But looking forward, prices will probably be falling for the OFHEO index in Q3 (to be released Thursday, November 29) and the Fed would then also show declining prices in the Q3 Flow of Funds report. Based on some rough estimates, it appears that over $100 Billion of the $189 Billion in household real estate asset increases in Q2 was from the estimated price increase. With declining prices in Q3, household equity will probably fall significantly - putting a big dent in Harney's argument.