Thursday, August 7, 2014

A recession without a boom?

Simon Wren-Lewis in one of his latests posts dismisses the idea that the pre-recession UK economy was in an unsustainable debt-fueled boom. The argument, which I have made before in earlier posts, is that focusing only on the path of debt can give a misleading picture of the sustainability of growth or spending. Wren-Lewis presents data for the UK showing that the large increase in debt in the UK prior to 2007 was matched by an increase in the value of the assets that UK households held so that net wealth was indeed increasing (despite the increase in borrowing). The role of housing prices is key to understand these developments. Wren-Lewis argues that in the case of the UK, because of a combination of lower interest rates and limited supply, housing prices are trending upwards.

The simplest way to understand the argument is to think about two alternative scenarios when it comes to the purchase of housing services: renting versus borrowing to buy the house (no rent is paid in the future but mortgage payments will have to cover the cost of borrowing).

In the first scenario, there is no debt, in the second one there is an increase in debt. In what sense the presence of debt makes the second scenario or decision more unsustainable than the first one? One could argue that because of the ownership of the asset, there is the possibility of decreases in asset prices that could result in financial pressure and need to reduce other spending. Correct, but the same could be argued in the first scenario where a potential increase in the price of houses (and rents) will reduce the after-rent income for the household and will result in the same need to reduce other forms of spending (there will be no need to "deleverage" but the need to cut other expenditures will be identical). There are, of course, differences in terms of the ability to adjust to shocks because of the illiquidity of housing as an asset, but fundamentally there is a great deal of symmetry where ownership of the asset will give up potential returns if the asset price goes up but risks if it goes down while renting will exposes you to a risk of increases in rents if housing prices go up.

What can make the analysis more difficult is the heterogeneity across agents. Housing happens to be a special asset where it is mostly held to enjoy the services that it provides. In addition, we are all somehow exposed to fluctuations in housing prices because we either rent or buy the asset. This means that when we think about an individual scenario of renting we cannot forget that someone else owns the asset. This can in some cases reduce the aggregate risk as increases in prices benefit owners but hurt renters but the aggregate effects can depend on the characteristics of the individuals who are exposed to different risks. For example, while aggregate net wealth today in the US is higher than before the crisis (according to the US Federal Reserve flow of funds database), this fact hides strong difference across different households as there is evidence of increased wealth inequality during that period using alternative datasets. And this heterogeneity could have aggregate consequences because of differences in accessing to financing or spending patterns. But in this case the story is more complex than the notion of a country living beyond its means through a borrowing spree.

So increases in debt alone cannot be interpreted as a definite signal that the economy is in an unsustainable boom. But we might not rule out either that it might be a reflection of other economic trends that can put growth at risk. in particular when we take into account the heterogeneity of agents. Unfortunately the lack of available disaggregated data as well as solid and accepted theoretical models with heterogeneous agents make this analysis a lot more speculative.

Antonio Fatás