The limits of “borrowing to build”

Funding for building new council housing has always included borrowing, usually from the government’s Public Works Loans Board, at interest rates which are lower than commercial rates. The news that the government is ending the borrowing cap, i.e. the amount of borrowing which each council which owns housing stock is allowed, has been widely welcomed as a means of increasing new building. Estimates vary as to how many homes a year will be built by councils taking on more debt. The Ministry of Housing, Communities and Local Government (MHCLG) estimates that councils will be able to build an extra 10,000 homes a year. However, given that councils are losing somewhere in the region of 15,000 homes a year from right to buy (RTB) sales and demolitions, this would not even stop the annual decline in the number of council homes in England. So long as RTB remains in operation more than 15,000 extra homes a year would be required to increase stock numbers even marginally.

Much of the comment on lifting the cap takes no account of two factors:

• The level of existing debt which has to be serviced. At the end of the last financial year the combined debt of local authority housing revenue accounts (HRAs) in England was £26 billion. According to the Local Government Finance Statistics around 25% of annual income is eaten up by debt and interest charges.

• There is no consideration of who pays for the borrowing. Under the new council housing finance system debt servicing is paid for by tenants through their rent and service charges. If there is no central government grant available for new building then the cost falls entirely on the shoulders of existing tenants. The danger is that if councils take on more debt then the cost of servicing it will eat into the resources available to maintain existing housing stock.

Building new housing, of course, adds to the revenue stream of HRAs because they bring in more rent. Eventually the rent collected will pay off the debt, if a council repays the loan¹. But if the extra rent from new properties does not cover the cost of the new borrowing then the difference has to be covered by HRA revenue. This means that less money will be available for maintaining and renewing existing housing.

So, with this in mind, in this article we examine the impact of extra debt and the associated costs. We compare financial projections for building 200 new council homes in Swindon with three scenarios:

• Funding by borrowing alone²;

• £60,000 government grant per property and

• £80,000 government grant per property.

Under the New Labour government, until the 2008 crash there was no support for building new council housing. Councils were not allowed to bid for ‘social housing’ grant. Thereafter its National Affordable Homes Programme did introduce grant of around £60,000 per property though councils had to bid against housing associations for the money available³. The £80,000 grant figure was suggested by the DCLG (now the MHCLG) when the government announced last year that there would be any extra £2 billion available for “affordable housing”. It said that this amount could, for example, help to build 25,000 social rent homes with grant at £80,000 per property, though this was no commitment.

The elements of a plan for building new housing include the cost of building, the cost of borrowing (repayment of the loan and annual interest payments), any government grant and the rental income. The latter is dependent both on government policy (central government determines council rent increases) and inflation.

Of course, financial projections over 30 years are very broad brush estimates. The projections we have made are based on these assumptions:

• The average cost of building a home taken as £125,000.

• Borrowing from the PWLB over 30 years, with a fixed interest rate of 2.9% (the rate when this article was written).

• An equal annual repayment of the loan over 30 years;

• Starting rent estimated at £90 per week for 48 weeks (slightly above the current average rent for Swindon council homes);

• An annual inflation rate of 2% and hence rents increased by that amount⁴;

• No cost of land (i.e. council land will be used).

Apart from the fixed term interest rate all the other factors are variables which can and will change. Even the most sophisticated accountant cannot predict what will happen in five years never mind over 30. So these and any other financial projections are working hypotheses which have to be amended year by year. In real life variables change and impact on each other.

The main purpose of these projections is as an illustration of the process of “borrowing to build”, and a comparison of what happens with no grant, and two different amounts of grant. What you can see from these projections is that, in the case of borrowing only, the rental income is insufficient to cover the cost of borrowing within 30 years. In Table 1 the rental income does not cover the annual cost of repayment of the loan and interest charges until year 18. At the end of the 30 years when the debt payments end the rental income has not recouped the full cost of borrowing, being £2.7 million short.

The big difference that government grant makes can be seen in Tables 2 and 3. For both the examples of grant at £60,000 and £80,000, there is an annual surplus (rent income being more than repayment of loan and interest charges) in year 2. Over the course of the 30 years the overall surplus is £14 million and £20 million respectively.

In all 3 cases the simplifying assumption is that 200 homes will be ready to let by the end of the first year, though in reality the building will be staggered, so the rent income would be lower than shown. If the council has to buy land then the projections in Table 1 would be worse, though probably covered by the other two scenarios.

The fundamental problem with the first scenario is that, since rent does not climb above borrowing cost until Year 18, the cost has to be borne by the HRA. In other words money which would otherwise be spent on the upkeep of existing homes has to be diverted to payments to the PWLB. For the first six years that is more than £500,000 a year.

This would have an impact on capital spending which in Swindon is just over £15 million a year. This is investment in the housing stock, replacing key housing components such as kitchens, bathrooms, boilers, roofs etc. Under the Decent Homes Standard each of these components is given x number of years of useful life. If they are not replaced in good time then they would need to be repaired more often, driving up the cost of revenue spent on day to day maintenance. The loss of £500,000 could, for example, mean that more than 100 tenants whose kitchens need replacing would have to wait longer.

This is the fundamental problem with the idea that “borrowing to build” is sufficient to fund a large scale council building programme. Of course, the average cost of building will depend on the types of properties being built, the number of bedrooms, and will vary from town to town, just as council rents vary in different localities. If the average price of building was £100,000 then in the case of borrowing only, there would be an overall surplus of £4.5 million after 30 years. An annual surplus would be achieved in year 12 instead of 18, so the annual loss would have to be covered by the HRA for 11 years. The surpluses in the case of the grants of £60,000 and £80,000 would be correspondingly larger; nearly £22 million and £28 million respectively.

One thing that these projections don’t take into account is the loss of homes from right to buy (RTB). Each home sold represents rental income lost. Based on these projections just one home sold after three years (the current time for tenants to qualify) would mean the loss of around £150,000 in rent income up to year 30. Debate on RTB often centres on the number of homes lost but the question of the impact on HRA resources – lost rent – is often forgotten. That’s another reason why RTB should be ended.⁵

Conclusions

The financial projections we have shown are rough ones. The interest rate was current for the PWLB. The assumption in relation to inflation is obviously very broad brush but you cannot predict what inflation rates will be in the future. Fluctuations will increase and decrease rent projections. These caveats notwithstanding it is clear that government grant on the scale which was provided previously, and might be in the future, (the value of £60,000 has been eroded by ten years of inflation) enables investment in new homes to produce an annual surplus very quickly, increasing the revenue of HRAs for the upkeep of existing homes and providing scope for more building. In contrast, building based on borrowing alone, does not produce an annual surplus for a very long time and puts a strain on the resources of HRAs.

We have argued elsewhere that the so-called debt imposed in 2012 was bogus; the product of ‘creative accounting’ by the Treasury. We have put the case for it being written off/cancelled. Short of that there is also a good case for the current government cutting the debt in line with the significant loss of income that HRAs have suffered as a result of government policies such as the 4 year rent cut and the ‘enhanced right to buy’. This would significantly cut the annual outlay of councils on debt servicing or even write it off.

New borrowing combined with government grant at the levels we have shown would completely transform the ability of councils to begin building on a large scale once more, something that even the Tory Leader of the LGA, Gary Porter, has said is necessary. He pointed to the fact that the last time 300,000 homes were built in a year councils built 40% or more of them. “We have to get back to that”, he said. Building on that scale is impossible without a significant increase in central government grant. To build 100,000 council homes a year, with £80,000 grant, would require £8 billion a year. Whilst that sounds like a lot of money, compared to government spending overall it is not a large sum. It is a question of political priorities. Borrowing alone is no substitute for central government grant if we are to return to council house building programmes on a large scale.

Martin Wicks

Secretary, Swindon Tenants Campaign Group

November 2018

Download a PDF to see the Tables borrowingtobuild