7 December 2015

Why housing benefit is broken


Now the government has U-turned on plans to abolish tax credits, it should look at housing benefit for welfare savings—most housing benefit is a transfer to landlords and the remainder is an inefficient and distortionary intervention. It should abolish the £26bn system and use the money to fund tax cuts for low earners and a shallower Universal Credit withdrawal curve. Paradoxically, housing benefit is one of the causes of our housing crisis, rather than a solution.

Of the UK’s welfare schemes, housing benefit is the fourth most costly, after the state pension, disability benefits and tax credits. Reforms in 2012 and 2013 under the last government capped the total amount claimants could receive, and reduced the amount claimants were eligible for if they had a room judged ‘spare’ by the government’s definition. They also restricted recipients to claiming only for properties with rents at the 30th percentile for their area (i.e. lower than 70% or more of other rented properties). For example, claimants may now claim up to £260.64 a week if they live in a one-bedroom property, depending on their location, as well as their financial and personal circumstances.

The UK’s system is huge compared to most other rich countries, in both scope and generosity. In 2009, 18% of the Great British population was receiving cash allowances for rental support — more than anywhere else in the OECD. In only three other countries did more than 10% claim; in six countries fewer than 2% got these benefits. In 2005 the average claimant got around 18% of their rent covered — the second highest in the OECD, and again, most others were not marginally, but much lower.

The system has been improved, but only marginally; the two key problems inherent therein remain. These are firstly that, in our system of constricted housing supply, housing benefit merely adds to the competition over limited supply, driving up rents across the board. The bulk of the payments are effectively transfers to all landlords and homeowners. Secondly, the remaining portion, perhaps a quarter, which does lift the incomes of tenants, is distortionary in two key ways: it leads them to spend a higher proportion of their income on housing than would be optimal; and it slashes the incentives for those on low incomes to work, earn and save more.

The first of these is easy to understand in logical terms; and easy to see in the empirical data. The UK has a very restrictive planning system: this means that raising the price of housing (rents and house prices) does not lead to much more housebuilding. Builders are not holding back because there aren’t any more viable projects, but because getting planning permission on these is so difficult. This also means that giving (a subsection of) housing consumers a wad of housing vouchers does not lead to much extra housing being supplied. It’s just more money chasing the same amount of goods, which drives prices — i.e. rents — up. No one extra gets a home, we just shuffle everyone around; those claiming housing benefit rent houses that would otherwise be rented to non-claimants, all of it at a higher rent.

The published academic research finds the same result: higher housing benefit is mostly cancelled out by higher rents, and does not make housing more affordable. In Finland market rents faced by affected households rose €0.60-€0.70 for every €1 hike to the benefit in a 2002 reform. In France, a housing allowance system similar to the UK’s delivered €0.78 of every additional €1 spent in benefits to the landlord in the form of higher rents — only €0.22 went to the tenant. The UK reforms in the 1990s delivered similar results, as did other experiments around the world, according to a Melbourne Institute/OECD review.

So a large chunk of the benefits are a handout to landlords, a class already benefited by restrictive building regulations, as well as the recent welter of interventions centred on Help to Buy. But the remainder is little better.

Prices of goods reflect the costs of the ingredients needed to make them — which is in turn a measure of other uses they could be put to. When we tax or subsidise a good, without a special reason (like externalities), we ‘distort’ the ‘signal’ contained in the price — we debase the informational content of the price, and people do too much or little of it, relative to the socially optimal outcome.

For example, if we subsidise roads but not railways, we will divert too much steel, cement, tarmac, and most importantly labour power to building and not enough to railways. Similarly, if we subsidise housing by giving people vouchers that work only on it, rather than say, cash, people will shift their budgets towards it and away from travel, clothing, food, and entertainment. The real costs to society of their consuming a given property are still there, but they are not bearing them fully.

Much of the work on housing benefit notes this distortionary ‘overconsumption’ effect. It will not simply be size and quality, but location. Since housing is now effectively cheaper for benefit claimants, they will shift toward more central and prime locations, even when others with similar total incomes (but less through benefits) may value those areas more.

This is compounded by the fact that claimants are paid by the going rents in area — incentivising them to live in as expensive an area as possible. It’s like a car benefits system that paid out 10% of the price of any car — people will game the system as much as possible below the upper bound cap.

All of these distortionary effects worsen the housing crisis by matching people with less appropriate properties in less appropriate areas, and getting less efficient use out of our housing stock, even given the tightness of existing housing regulations. The effects may even be part of the cause of the worsening regional inequality and North-South divide in the UK, by pushing people to find the most expensive places to live, invariably located in London.

Scrapping housing benefit entirely would get the exchequer £26bn. Assuming at least 50% of this comes from landlords, then former recipients are made at most £13bn worse off in aggregate — likely considerably less.

There are numerous ways of compensating low earners in better ways: for example it would cost about £14bn to raise the threshold at which workers start paying National Insurance Contributions to the full-time minimum wage level (£251.25 a week or around £13,000 a year). We could also make the rate at which Universal Credit is withdrawn less steep, or beef up tax credits. Any of these would strengthen work incentives and efficiency, and enhance the progressivity of the overall tax and benefits system, relative to the current regime.

Housing benefit is a broken system. Most of the vast sums spent on it go into landowner pockets, instead of alleviating burdens on tenants. By misallocating property & incentivising moving to the most expensive areas they distort the housing market, reducing efficiency and potentially worsening the housing crisis and the North-South divide. We should not stop at simply tinkering with it: we should abolish it outright, and use some of the savings to compensate those on the bottom end who’d lose out, with one of our more efficient tools.

Ben Southwood is Head of Research at the Adam Smith Institute.