Last week the Fed released its Flow of Funds (FOF) data for all of 2005. The FOF data keeps track of changes in various types of assets and liabilities for each sector (and many major industries) of the economy.
One piece of the FOF data that I find particularly interesting is the picture that it paints of household finances in 2005. The following table displays the annual changes in the assets and liabilities of US households over the last several years.
Sources: FOF tables F.100 and B.100. Note that this data is actually for households and nonprofit organizations. Nonprofit assets are estimated to comprise about 6% of the total combined category.
A couple of aspects of this data strike me as interesting. First of all, for the seventh year in a row, US households aquired more financial liabilities than they did financial assets. Naturally, this is due to the fact that much of the debt taken on by households was taken on to buy real estate. The escalating price of real estate meant that in 2005 the gap between the financial assets that households aquired and the financial liabilities they incurred was by far the largest ever.
It’s also interesting to note that households (and their pension funds) were net sellers of equities during 2005. Unlike in 2000, when the value of equities owned by households fell because of the fall in the stock market, in 2005 the stock market actually rose slightly, so the decline in the value of equities owned by households must be due to fact that households were net sellers of stocks during the year.
Putting this point together with the previous point means that, over the past few years, US housholds have been shifting the composition of their assets substantially away from financial assets – especially equities – and toward real estate. The following chart illustrates how households have altered their asset porfolio over the past 40 years.
Some obvious features of this picture are the general trend toward holding more wealth in the form of equities (this includes mutual funds and pension account balances), and the large spike in equities as a fraction of household assets that happened in the second half of the 1990s as a result of the great stock market boom. But over the past few years the dominant trend has been the rapid growth in real estate as a share of household wealth – to levels higher than at any time in recent history. Households even have far more of their wealth tied up in their house than they did in the 1960s or 70s, when stock market investment was almost unheard of for most households.
Note that this picture vastly understates the role of real estate in the typical family’s assets, since the huge majority of equities are owned by only the very richest Americans; the typical household actually has just a few thousand dollars in equities, so 90% or more of the their wealth is typically in the form of real estate.
An interesting question is why households have elected to put an increasing share of their wealth into houses in recent years. I’m not sure I have a good answer to that question. But put this all together, and it provides a reasonable explanation for the evidence that suggests that consumer spending is quite a bit more sensitive to changes in real estate wealth than it is to changes in stock market wealth. The next few years may be an interesting test of this relationship.