Is the government fudging the numbers when it comes to housing promises?

The government has sometimes given the impression that if the target is met (a very big if), then a big increase in affordable housing will inevitably flow from that via Section 106.

But all the evidence suggests that this is the wrong way around and that it can only hope to come close to 1.5 million homes if a significant proportion of them are affordable.

To begin with, though, a generation is usually taken to mean 25 to 30 years. That takes us back to 1995-96, when output had reached a record high of almost 75,000 social and affordable homes in the wake of extra investment during the housing market crash.

The rest of the 20th century saw Labour cutting investment after 1997 in line with its pledge to stick to draconian Tory spending plans in its first two years in office before funding increasingly significant increases in its second and third terms.

“When we adjust to 2023-24 prices, a very different picture emerges. The peak was still in 2009-10, but investment would have to rise by 80% from 2023-24 levels to match it”

After 2010, the Conservative-led coalition slashed spending and ceased funding social rent altogether. Investment recovered after the Brexit vote in 2016 and social rent was rehabilitated, but on nothing like the same scale as before.

The graph below shows annual outturn investment by Homes England (and its predecessors) and the Greater London Authority in new affordable homes over the period. In cash terms, the peak for investment was 2009-10, and 2023-24 was the next biggest. Add the extra £800m the chancellor has found down the back of the sofa for 2025-26 and we arguably already have “the biggest boost”.

But that takes no account of inflation and would be completely misleading in the same way as government claims to be spending “record” amounts on things. 

With thanks to Gillian Young, an editor of the UK Housing Review, when we adjust to 2023-24 prices, a very different picture emerges. The peak was still in 2009-10, but investment would have to rise by 80% from 2023-24 levels to match it.

Quite a challenge for the Spending Review in June.

Another way of thinking about investment is to compare different Spending Review periods. The first of these in 1998 covered the three years from 1999-2000 until 2001-02. Spending fell by 14% in real terms compared with the previous three.

The next four Labour spending reviews in 2000, 2002, 2004 and 2007 all saw large increases in investment (it almost doubled in real terms as a result of the 2002 review, but this was from a low base).

By complete contrast, the coalition’s spending review in 2010 saw investment fall by 61% in real terms over the next five years compared with the previous five.

We could also look at performance in different parliaments. On this basis, “the biggest boost” achieved over the period was by the Labour government between 2005 and 2010. The current one would have to double investment in real terms by 2029 to match that.

“Social rent numbers would have to rise threefold on 2023-24 levels to match those achieved at the end of the last Labour government and almost fivefold to match the generational peak of 1995-96”

However, this does not tell the full story either. These real terms figures reflect increases in general inflation, not the larger increases in land and construction prices that mean the current Affordable Homes Programme (AHP) is set to deliver far fewer than the 180,000 homes originally expected.  

The figures also show total government grant towards affordable housing, and so do not include private finance raised by social landlords or contributions from developers via Section 106 agreements.

So perhaps a better way to look at it is to compare completions of social and affordable homes, which inevitably lag behind investment.

That is what is clearly implied in the government’s response to a housing committee report published earlier this month. This recasts Ms Rayner’s “biggest boost” as “the biggest increase in social and affordable housebuilding in a generation”.

This graph shows what has happened to completions over the past 30 years from all sources, including grant funding, Section 106 and social landlords’ own resources.

Here, too, the government could have a misleading (and weaselly) basis to claim that it has generated the biggest increase without actually doing very much. Total output of social and affordable homes would have to rise by just 6% on 2023-24 levels to beat the recent high of 2014-15, or 20% to beat the 1995-96 peak.

But that would of course ignore the way that the AHP shifted away from social rent (in red) to affordable rent and affordable and shared ownership under the Conservatives between 2010 and 2024.

The Labour government has been clear that it wants to see social rent take priority in the remainder of the existing AHP and in the new one. If that means ‘the biggest increase in social AND affordable housebuilding’, there is much more work to do. The social rent numbers would have to rise threefold on 2023-24 levels to match those achieved at the end of the last Labour government and almost fivefold to match the generational peak of 1995-96.

And annual output of affordable homes would have to rise by 46% over this parliament to match the increase achieved by Labour in its third term between 2005 and 2010.

Even those highs, of course, would not be sufficient to generate the almost 500,000 affordable homes that analysis by Savills suggests could be needed to add to market output to meet the 1.5 million-homes target over the five years of this parliament.

And they come nowhere near the 90,000 social rent homes a year now required to meet housing need, a figure endorsed by housing organisations, independent researchers and the all-party Housing, Communities and Local Government Committee.

But any of these generational increases would be a pretty good start.

Jules Birch, columnist, Inside Housing