Falling house prices are viewed by many as a major contributor to the economic slowdown in the US. In Europe, a long run of years characterised by rapidly rising house prices appears to have come to and end in Ireland, Spain and in the UK. Lower house prices affect aggregate demand through two channels – investment in housing and private consumption demand. This blog deals with the effect of changes in housing wealth on private consumption. It uses a little algebraic notation as part of a scientific experiment to determine the impact of symbol use on blog readership.
The effect of changes in the prices of existing homes on investment in home improvements and on the construction of new homes is not considered, except for noting the differences between the current size of the residential construction sectors across the four countries just mentioned. In the US, construction accounted for about 5.0 percent of GDP in 2006. The corresponding 2006 figures for Ireland, Spain and the UK were, respectively, 9.9 percent, 12.2 percent and 5.4 percent of GDP.
On average, you live in the house you own
The bold statement that is the title of this blog was put to me about ten years ago by Mervyn King, now Governor of the Bank of England, then Chief Economist of the Bank of England, shortly after I joined the Monetary Policy Committee of the Bank of England as an External Member in June 1997. Like most bold statements, the assertion in the title of this blog is not literally correct, although the pithy statement trumps the literal truth as a vehicle for getting the message across. A more precise statement would be that an increase in house prices does not make you better off. The argument is simple and applies to any consumer durable.
Let’s consider coconuts instead of houses. When does an increase in the price of coconuts make you better off? Answer: when you are a net exporter of coconuts, that is, when your endowment of coconuts, E, say, exceeds your consumption of coconuts, C, say. If P is the price of a coconut, the amount, G, by which you are better off when the price of coconuts increases by an amount ΔP, is given by
G = (E – C) x ΔP .
A net exporter of coconuts is better off when the price of coconuts increases, a net importer of coconuts is worse off. Someone who is just self-sufficient in coconuts is neither worse off nor better off.
Houses are no different from coconuts in this regard. Assume for the moment that the price of a house (strictly of a unit of housing) equals its fundamental value, F – the present discounted value of the future rents it generates. These rents can either be the market rents earned by a commercial owner of property-to-let, or the imputed rental value of owner-occupied housing. So we have:
P = F .
The consumption value of a unit of housing is also the present discounted value of current and future rents. When the price of residential property increases, households-consumers whose endowment of housing exceeds their consumption of housing are better off. Those for whom the endowment of housing is less than their consumption are worse off. The next step is the assertion that, for a typical household-consumer, the endowment of housing equals the consumption of housing:
E = C .
It follows that, since the representative household-consumer is self-sufficient in housing, a change in home prices does not make it better or worse off:
G = 0 .
As long as your endowment is positive, your wealth obviously increases when the house price increases. However, an increase in house prices means that the present discounted value of future rents has increased. As a consumer of housing services, now and in the future, you are therefore worse off. On average, in a country like the UK, people consume the housing services they own. Hence an increase in house prices does not make them better off. For financial assets like equity there is no corresponding “present discounted value of future equity services consumption” whose cost increases whenever the value of equity goes up. An increase in stock market values therefore unambiguously makes you better off.
But as regards house prices, regardless of whether a change in price is due to a change in risk-free discount rates, in risk premia or in expected future rents, you are neither better off nor worse off as a result of that price change, if you consume, now and in the future, the same contingent sequence of housing services whose present discounted value is part of the wealth you own. In that case, despite the increase in your housing wealth, once you have paid for the consumption of your initial contingent sequence of housing services, there will be nothing left to spend on anything else.
Instead of the price a house, P, equalling its fundamental value, it could be equal to the sum of its fundamental value, F, and the value contributed by a speculative bubble, B:
P = F + B .
The consumption value of the house, PC continues to be equal to the present discounted value of current and future rents, that is,
PC = F .
It follows that the net financial benefit or gain you have from a change in house prices can be written as
G = (E – C) x ΔF + E x ΔB .
So, making again the assumption that E = C, the net gain from an increase in house prices is given by
G = E x ΔB .
Ironically, an increase in house prices therefore only makes you better off if there is a (positive) bubble.
Some households-consumers have an endowment of housing that is smaller than their consumption of housing (call them renters). Others own more housing than they consume (call them landlords). Clearly, landlords-homeowners are better off when the price of houses increases, while renters are worse off. If the marginal propensities to spend of these two categories of households-consumers are different, such a redistribution will not be neutral as regards its effect on aggregate consumption. I have no empirical information on whether renters have higher or lower marginal propensities to spend than landlords.
Likewise, some current home owners may be planning to ‘trade down’ later in life, for instance when the family home gets replaced by a smaller property when the children leave home, following retirement or following the death of one’s spouse. For them the fundamental value of their endowment exceeds the present discounted value of their current and future planned consumption of housing services. Against that, there are also persons planning to trade up in the housing market.
Residential property in a country could be owned by households-consumers not resident in the country, in which case the fundamental value of the housing stock located in the UK and owned by UK households-consumers is less than the present discounted value of future rents from the stock of houses located in the UK. An increase in the price of UK property would make UK households-consumers worse off in this case. I have no information on how much UK residential property is owned by foreign households-consumers.
Consumers of housing services may be more myopic and/or may discount future expenditure on rent at a higher rate that the discount rates that are reflected in house prices. Home owners may fail to recognize the opportunity cost of the owner-occupied housing services that they are consuming. There may be a variety of other irrationalities and behavioural idiosyncrasies causing the consumers of household services to respond differently from the owners of homes to changes in the present discounted value of current and future rents.
What do economic models assume about the effect of housing wealth on consumption demand?
Most econometric models I am familiar with treat housing wealth like equity as a determinant of household consumption. They forget to allow for the fact that households consume housing services (for which they pay or impute rent) and that with properly functioning markets, the value of housing wealth in the consumption function would be cancelled out exactly by the present discounted value of current and future rents.
An example is the FRB/US model. It is used frequently by participants in the debate on the implication of developments in the US housing market for US consumer demand. A recent example is Frederic S. Mishkin, a member of the Federal Reserve Board, in a paper “Housing and the Monetary Transmission Mechanism”, NBER Working Paper No. 13518, October 2007. The FRB/US model a-priori constrains the wealth effects of housing wealth and other financial wealth to be the same. The long-run marginal propensity to consume out of non-human wealth (including housing wealth) is 0.038, that is, 3.8 percent. In several simulations, Mishkin increases the value of the long-run marginal propensity to consume out of housing wealth to 0.076, that is, 7.6 percent, while keeping the long-run marginal propensity to consume out of non-housing financial wealth at 0.038.
The argument for an effect of housing wealth on consumption over and above the pure wealth effect, is that housing wealth is collateralisable. Households-consumers can borrow against the equity in their homes and use this to finance consumption. If they are otherwise liquidity-constrained or credit-constrained, a boost to housing wealth would boost consumption by more than the pure wealth effect.
Housing wealth is, however, not the only form of financial wealth that allows one to overcome liquidity constraints or credit constraints. Many other financial assets can be used as collateral. In fact, liquid financial assets like stocks and bonds tend to make better collateral than housing equity. Indeed, one can finance consumption spending simply by running down holdings of liquid financial assets; unlike illiquid housing wealth one does not have to use them as collateral for a loan in order to turn them into current spending power.
You can, of course, get a larger aggregate liquidity or collateral effect from housing wealth than from other financial wealth, if housing wealth is owned to a greater degree by liquidity-constrained households than other financial wealth. To the extent that the ‘other’ financial wealth is liquid, it is obviously true that the owners of this ‘other’ financial wealth cannot be more liquidity constrained than the owners of residential property. This, however, does not constitute evidence that owners of residential housing are liquidity constrained. It could be the case that neither the owners of residential property nor the owners of ‘other’ financial assets are liquidity-constrained. On balance, however, it seems likely that at least some owners of residential property are liquidity constrained while, almost by definition, no direct owners of liquid financial instruments are liquidity constrained.
What this implies is that the ‘collateral’ or ‘liquidity’ effect of a change in house prices on consumer demand is likely to be larger than the ‘collateral’ or ‘liquidity’ effect of a change in other financial asset prices of equal magnitude. This does not mean, however, that the total effect on consumption demand of a change in house prices will be larger than that of an equal change in the value of other financial assets. This is because the total effect is the sum of the pure wealth effect and the liquidity/collateral effect, and the pure wealth effect of a change in house prices on consumption demand is zero.
If the argument in this blog is correct, the FRB/US model’s approach to the consumption effect of housing wealth is questionable. The pure wealth effect of housing wealth on consumption is zero (not 0.038). Mishkin then makes the further, untested, assumption, that the collateral or liquidity effect of housing wealth on consumption is greater (0.038) than the collateral or liquidity effect of ‘other’ financial wealth (0.000).
The arguments of this blog lead to the conclusion that the proposition that consumption is affected more by changes in housing wealth than by changes in other financial wealth is questionable, and that there is a distinct possibility that the net effect of changes in housing wealth on private consumption may be small or even negligible (if home owners are not liquidity- or collateral-constrained).
So, when considering the impact of housing price declines on aggregate demand, focus on the investment and home-improvement side, not on the consumption effects.