The long-term cost of Help to Buy

Originally posted on June 13 on my blog for Inside Housing. 

Whatever you love it or hate it, Thursday’s report from the National Audit Office (NAO) will probably not do much to change your mind about Help to Buy.

If you think that the equity loan scheme first launched in 2013 has boosted housebuilding and helped more people to buy their first home, you will find evidence to support that view: new-build  property sales increased from 61,000 a year in 2012/13 to 104,000 in 2017/18; and around 81% of people using the scheme have been first-time buyers.

If you think the scheme has mainly benefited housebuilders and the benefits for buyers have been more limited, you’ll find backing for that too: 63% of borrowers could have afforded to buy anyway; many of them have used the scheme to buy a bigger house than they could previously have afforded; and 10% of buyers had incomes higher than the £80,000 (£90,000 in London) limit for eligibility for shared ownership.

The report does reject one common allegation made against Help to Buy by estimating that homes sold under the scheme have cost just 1% more than similar new-build homes. Previous estimates ranging from 5% to 20% have not compared similar properties, says the NAO.

However, that is just part of a much bigger new-build premium (the difference between prices of new and second-hand homes) and the NAO seems to accept the high figure of a premium of 15-20% as a given rather than the product of market conditions that Help to Buy helped to create.

So, take your pick, this report won’t change your mind that much. It takes up the themes of previous work by the NAO and its conclusions are informed by the government’s own evaluation.

Clearly something must account for the extraordinary increase in profits, bonuses and the rest of the big housebuilders and the report presents without comment an extraordinary chart showing their share prices have quintupled relative to the rest of the FTSE 2050 since 2008.

However, as I’ve argued before, Help to Buy is only part of that story. Ultra-low interest rates and a whole range of other indirect government subsidies and cuts to red tape have also played their part in generating overwhelmingly favourable business environment for housebuilders.

A key issue looming here is what happens when the Help to Buy drugs are withdrawn in 2023: is the government right to argue it has given the market enough warning to suggest; or will housebuilders suffer the withdrawal symptoms implied by its acceptance that the industry will have to be ‘weaned off’ the scheme?

The NAO’s job is to determine whether spending programmes offer value for public money. On this central point the report concludes that up to now the answer is yes but in the longer term it is too early to say.

The government has put ‘reasonable arrangements in place to benefit from increasing property prices’ but if the market turns down ‘the taxpayer could lose out significantly’ and Help to Buy owners could in turn be trapped in negative equity exacerbated by that new-build premium.

This much has been clear from the beginning – on current trends the government is on course to make a profit from Help to Buy (and it has previously discussed cashing in on this by selling off the loan book) but things could look very different in a market.

The new element for me, though, was an appreciation of the significance of Help to Buy relative to the wider ambitions of Homes England, the government agency with a brief to disrupt the housing market and make strategic investments like the one recently announced in the joint venture between Urban Splash and Sekisui House.

The current Help to Buy scheme ends in 2021, when it will be succeeded by a more limited version for the next two years. On current projections, the Ministry for Communities Housing and Local Government (MHCLG) expects to have loaned £29 bn by March 2023.

The rate of return on this investment depends partly on what happens to house prices and partly on when borrowers redeem their loans.

The report quotes estimates by Homes England that a 20% fall in house prices would cut the portfolio value by £2bn (or 29%) while a 30% fall would cut the value by £5bn (47%).

The NAO adds comfortingly that:

‘These are falls in portfolio book value which the Department does not consider would turn into overall cash losses.’

In the longer term, Homes England currently forecasts that total redemptions will equal the amount loaned by 2031/32 and that by the time all loans are redeemed in 2048 ‘the scheme will have achieved a positive return on investment’.

If the fact that the impact of Help to Buy will still be felt in 28 years’ time is one indication of the long-term impact of the scheme, another is that the loan book was already worth £8.3 bn by March 2018 and that this represented 72% of Homes England’s total assets of £11.6 bn.

And this in turn raises the issue of what the government could have done with the money instead – the opportunity cost.

The NAO points out that Help to Buy represented 29% of the MHCLG’s total expenditure in 2017/18 and that:

‘Although money is to be recovered over time, a large proportion will be unavailable for other housing schemes or departmental priorities during the scheme and for a period after its end.’

Equity loans are, of course, financial instruments, so it’s not necessarily the case that the money could simply have been redirected into the affordable housing budget as grant.

However, this raises a final point about the long-term significance of a scheme dreamt up by George Osborne at the height of his austerity drive in the public finances.

As the NAO report shows, a programme of equity loans for housing can offer long-term value for money once that borrowing is repaid (which may take until 2048) provided the government is prepared to accept the risk that house prices can go down instead of up.

If that bet pays off, then perhaps by 2048 we’ll see it as one worth that was worth taking that boosted housebuilding, gave tens of thousands of first-time buyers a way into the market and made a profit for the taxpayer into the bargain. If It goes wrong, perhaps not so much.

But what if there were an alternative? If only there were a way for government to borrow to invest in housing pretty much risk-free, for the loans to be repaid from rents over 25 to 30 years, and for the homes themselves to be retained for the public benefit.

From the perspective of 2048, we might just see that as a better deal.