Economist – Conservatives to Shift Away from Emphasis on 300k National Housing Aspiration

No No No, does Gove want to repeat EVERY mistake of Eric Pickles? Mulheirn is a charlatan that leave housing experts head in hands, he doesn’t account for teh vast increase in second homes and holiday lets, he doesn’t even make a distinction between houses and households, doesn’t account for household suppression, or even understand the concept of local housing markets as he aggregates every statistic at a national level. He didn’t even understand (a mistake the economist also makes) that net completions data includes flat conversions. He has been debunked again and again. He wouldnt pass a Housing Statistics 101 class at first year

He is to Advocado Nimbys what Piers Corbyn is to ant-vaxers, and Alex Jones is to MAGAs. Their fake news acolyte.

So its Tory Policy now to let house prices to rise by 15% through restricting supply is it? Remember this is a compound rate so over a generation, compared to Germany, housing consumes twice the proportion of real wages as in England. This is at a time when over the last 5 years their has been a flood of peer reviewed studies which are creating a consensus in urban economics that their is a very clear correlation with house price affordability and zoning restrictions.

I should note it was Jenryks spads that were briefing this way. They are out on their arses now.

Economist


In 1975, in
 her first conference speech as Conservative leader, Margaret Thatcher pledged to build “a property-owning democracy”. More than 45 years later homeowners are still more likely to back the Tories than their rivals, and so the government would like there to be more of them. In 2017 it set an ambitious target of increasing the housing stock by 300,000 units a year in England. To help hit that target, in 2020 Boris Johnson, the prime minister, promised to simplify planning and weaken locals’ rights to object to development.

But housing and planning present the Conservatives with a dilemma: although they want more homeowners, they do not want to annoy existing ones. A by-election loss in June to the Liberal Democrats in the constituency of Chesham and Amersham, prosperous commuter towns, was chalked up in part to local anger over planned development. It spooked the party leadership. Backbenchers, too, are growing nervous. Political insiders now expect a shift in emphasis at the Conservative Party conference, which starts on October 3rd, away from building new homes and the 300,000 annual target, and towards increasing taxpayer support to enable first-time buyers to take on larger mortgages.

British homes are expensive. Since 1995 prices have gone up by 170% in real (ie, inflation-adjusted) terms, one of the fastest increases among rich economies (see chart). The median house in England and Wales cost five times the median salary in 2002, but closer to eight times by 2020. Home ownership, which had been rising for decades, peaked in the early 2000s). Over the past two decades a consensus has emerged among politicians and policymakers that the main reason has been too little supply. But since the mid-2010s Ian Mulheirn, an economist who now works at the Tony Blair Institute for Global Change, has argued that the main causes of higher house prices have been falling real interest rates and looser credit.

For a long time, Mr Mulheirn seemed a voice in the wilderness. But now, it seems, the government is listening. Michael Gove, who became the minister responsible for planning and housing in September’s reshuffle, has reportedly taken to telling colleagues that only around 15% of the growth in house prices of the past two decades can be explained by lack of supply. Nick Boles, a minister for planning in the early 2010s, an ally of Mr Gove and long almost evangelical on the need for planning reform, has changed his view too.

Mr Mulheirn provides cover for the new direction. He has long argued that there is little evidence of an undersupply of housing. He cautions against reliance on house-building volumes: what matters, he says, are total net additions to the dwelling stock. The conversion of a three-storey house into three flats is materially the same as building two new flats, but is not captured by a focus on new building. Between 1996 and 2018, the dwelling stock in England grew by an annual average of 168,000 while the number of households grew by an average of 147,000. The result was that the net surplus of dwellings rose from around 660,000 to around 1.1m. And although the price of houses as an asset has indeed soared, he argues that the price of housing as a service has not. Rents have risen more slowly than median household incomes since 1996.

On this view—which, it now seems, the government shares—the main cause of higher prices is the role of housing as an asset. Owning a home produces an implicit income: the saving that would otherwise have been spent on rent. As with other assets that provide an income stream, such as stocks or bonds, the value of that income is determined by the interest rates available elsewhere. According to this logic, house prices are high for the same reason that bond and stock prices are high: because interest rates have collapsed and credit has become looser. Mr Mulheirn does not entirely reject the notion that more supply would lower prices, but estimates that two decades of adding 300,000 units a year would reduce prices by only around 10%.

Castles in the air

This theory is not without critics. The rental data on which it is based is of low quality. Household formation does not just influence house prices; it is also influenced by them. For example, high prices may nudge young adults to stay living with their parents for longer. Real interest rates are low pretty much everywhere, but housing markets vary by country and region. Rental yields are considerably lower in London than in northern England.

In any case, even if global factors explain most of the past two decades’ rise in house prices, increasing stock in the places where people most want to live would be both welfare-enhancing and good for productivity. And encouraging young people to take on more debt need not be the sole way to improve their living conditions. More social housing could be built, and conditions in the private rental sector improved. The taxation of housing could be reformed, the better to reflect rising values. But for a government committed to increasing home ownership, none of this would help hit its central target. ■

Let pick apart the central economic fallacy

On this view—which, it now seems, the government shares—the main cause of higher prices is the role of housing as an asset. Owning a home produces an implicit income: the saving that would otherwise have been spent on rent. As with other assets that provide an income stream, such as stocks or bonds, the value of that income is determined by the interest rates available elsewhere. According to this logic, house prices are high for the same reason that bond and stock prices are high: because interest rates have collapsed and credit has become looser.

All of this is correct, by itself, however there are 4 mistakes of omission.

1) It assumes the supply elasticity curve for house prices is horizontal – it is not a fantastic assumption and ignores all the evidence.

2) It assumes the increase of capitilisation of annual income streams for housing are SOLEY to do with capitalisation of interest rates and those interest rates having nothing to do with supply. The error here is that markets always capitise returns on both profits and rents. And he assumes because services don’t have rents neither do house prices. Land markets are unique in that they include profits AND rents.

3) You never get financialization of an asset unless that asset is scarce. This is related to point 3. You don’t get garages selling petrol for 5 pounds a litre of petrol is freely flowing to the stations. Ask yourself why housing assets have become financialised, and why for example cornflakes have not.

4) It ignore the key feature of housing markets. If a person buys a new house it frees up an old house to supply. This doesn’t happen with financial assets for example. This creates the well documented ‘filtering effect’ where newly formed households buy newly available existing units on the lower cost margin of supply.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s