The way we calculate our housing need- that is, how many houses each local authority should be building each year- has become a very strange world indeed. If gov.uk did wormholes then looking into what is known as ‘the standard formula’ is definitely one of them. As the country awaits an overdue planning bill and £24 billion of public money is to be spent ‘unlocking’ more than a million new homes between now and 2026, why is the party of ‘Build Back Better’ using household projection forecasts from 2014 and affordability ratios from 1998 to tackle a housing crisis in 2021?
A few years ago I went to NASA’s Kennedy Space Centre and like most visitors left with a piece of merchandise from the shop on the way out. It’s pictured here and it’s a glass that describes rocket science; the various different calculations that eventually led to humans being able to launch a rocket into space. I often look at it, understanding absolutely nothing (quite clearly not the laws of gravity having dropped it twice now), and think about the background work that had to happen to produce a final set of equations so absolute that they take every single thing into account to arrive at a perfect point of understanding.
Fast forward to 2021 and the glass happened to be sat on my desk when I first started digging into the reasons behind the housing crisis. It was the day I came across the standard formula; an equation devised by the government and the development industry to work out how many houses should be built in each local authority. Introduced in 2017 the standard formula removed powers from the authorities to decide how many homes it needed to build annually and centralised it to make sure the national house-building target would be met.
This is it:
So, spying the NASA glass out of the corner of my eye, I started looking into it.
Now I think its important to say I wasn’t expecting to find the missing unification theory nor is there a direct comparison between the rocket scientists at NASA and the UK Government. But I did have a basic assumption that some smart people in Whitehall, much like the scientists at NASA who had had to grapple with the various forces of nature before pressing go on a launch, had done some form of modelling that took various key housing market forces or influences into account to come up with the equation hailed as the solution to the housing crisis.
However, it turns out the standard formula is anything but rocket science.
The birth of the formula started with the emergence of one figure; a target of 300,000 new homes to be built annually for the term of the parliament. So it’s important to remember that the starting point (300,000) was always the desired outcome from any equation.
It’s not a number that is without evidence. Originally, 282,000 homes a year was put forward as a potential build target in a 2004 report commissioned under the Labour government called ‘The Barker Review’. This is a 155 page investigation into the housing market which details the level and type of house building that needed to happen from then until 2016 to deal with the existing shortage already built up, to provide for the forecasted population growth and to achieve a real price increase of just 1.1% annually rather than the 2.7% at the time (and 2.4% over the previous 30 years). More recently the author of that report, Dame Kate Barker, once again re-iterated the need for a similar number of homes- around 300,000- to be built annually in her evidence to the 2017 House of Lords Select Committee on housing. Then in 2018, research by Professor Glen Bramley at Heriot-Watt University (commissioned by the National Housing Federation (NHF) and Crisis) further identified a need for 340,000 homes each year in England to 2031.
So, as headline figure, 300,000 does have some historical weight behind it and other reports- both governmental and external have come up with similar figures for housing need.
Helping more people onto the housing ladder is always a cornerstone of any Conservative manifesto. It makes sense; statistically homeowners are more likely to vote Conservative, so turning ‘Generation Rent into Generation Buy’ could also be seen as turning ‘Generation any other party into Generation Tory’. It is this basic foundational requirement for the party that will always dictate the direction for any housing policy. With the highest build numbers since the Second World War being promised, the only question left for the government to decide was where these houses should be built. That’s where the standard formula comes in.
Essentially the formula is a convenient way of divvying up where to build the annual target and which areas take the brunt. Although local authorities do the final calculation themselves, the equation is already written for them and all they have to do is fill in the blanks with some localised data and voila; your building target is revealed.
‘Oven ready’ some might say.
Lets break the equation down- using Cornwall data- to illustrate how it works. Stick with me; it really isn’t rocket science.
The foundation of any housing need assumption is always the population data. The Office of National Statistics (ONS) produces population forecasts nationally and for the local districts and authorities across England. From these population forecasts it derives the ‘household formation forecasts’. These are essentially how many new households the ONS expect to form, based on the growth in the population data. This household population data is used as the starting point for any given authority to know how many houses they need to build.
If we replicate Cornwall Council’s current development phase, which runs from 2010 to 2030, the household formation data forecasts that 41,978 new households are due to be formed over the 20-year period. That works out at an average of 2099 per year.
So 2099 is the baseline figure that the standard formula manipulates to create a new ‘correct’ figure of the properties it says needs to be built.
Once we have our baseline we can look at the local affordability ratio. This is a measure of how affordable houses are in any area. It’s based on ‘affordable’ being a house that costs 4 times the average annual income. An area with an affordability ratio over 4 must use the formula to make the ‘adjustment’ to the baseline figure as this is where, because they are the least affordable areas, that housing must be needed the most.
Out of 335 localities the government produce an affordability ratio for, only 5 have an affordability ratio of or below 4, so about 1.49%. That means 98.51% of the country has housing prices well over what is deemed acceptable. This isn’t surprising though when you realise that 4 is the affordability ratio from the late 1990’s when over 95% of the country did have an affordability ratio of around 4. So it’s a strange figure to be using in 2021 when the house prices have gone up 170% in real (inflation-adjusted) terms since 1995 as opposed to the median annual income increasing by just 76% from £17,000 to £31,000 in the same time.
When we look at the government’s affordability ratios it gives us our second figure for Cornwall in 2010: 8.68. This means that the average house price in 2010 is just over 8.5 times the average salary. So now the formula, with Cornwall's data applied, looks like this:
The figure 1.17 means the excess affordability ratio is x1.17 higher than the acceptable affordability ratio.
This figure then gets multiplied by 0.25 meaning each time the ratio increases by 4 points the equation calculates that the baseline should increase by 25%.
The formula in this case deems that Cornwall Council must build 29% extra homes per annum than the household forecasts say there will be.
So we put those figures into the standard formula and the figure returned is the annual house building target of 2713 houses a year.
And that’s it. If you do that for all the local authorities you’ll get their share of the total 300,000 new homes needed per annum.
Get it? Well I get it. But I don’t ‘get it’. Why does a 100% increase in the affordability ratio automatically mean a 25% increase in the baseline housing stock required? There is no explanation or evidenced based reasoning for this anywhere.
In 2018 there was a major set back for the standard formula. When it was conceived, its architects achieved the outcome of 300,000 by manipulating the 2014 household formation forecasts. Unbeknownst to those mathematical wizards, at the very same time, changes were being made to the population forecast methodology within the ONS; the result of which was a decline in future population projections which therefore impacted the household projections and everything down the line from those leading pieces of data.
So when the government recalculated the housing need using the standard formula with the new 2016 ONS household forecast data it actually led to fewer houses needed than 300,000. The developers were not very happy about this and neither were the government. So they proposed to only use the 2014 data.
So that is now what happens. Out of date 2014 household projection forecasts and affordability ratios from a previous millennia are used as the basis for measuring and tackling the housing crisis in 2021.
Recognising the absurdity of not using their own up to date household forecasts the government proposed altering the formula in 2020 so it would once again return the magic number of 300,000 homes annually using the new 2016 data. But this would have resulted in such a uplift on homes built in various rural areas that in Cornwall it would’ve led to a doubling of how many houses needed to be built from a 2000+ baseline figure to over 4000+ annually. Local councils wrote to MPs saying it was impossible to justify or in fact deliver such targets and constituents, some of which were the MPs themselves, went berserk thinking there may be thousands more homes built at the end of their garden. The then housing minster Robert Jenrick realised this was definitely not a goer so the government relented and made the decision that the 20 top cities would automatically be assigned a 35% uplift on top of the result of the formula. However we still use to 2014 population figures because that’s the only way to get to 300k homes target annually.
We still await a new housing and planning bill after the 2020 white paper.
A Cornwall side note:
It is important to note that although I’ve used Cornwall figures as an example, Cornwall Council already had a local plan in place running from 2010 to 2030. It had been adopted in 2016, which was just before the formula was published. It was deemed ‘satisfactory’, with house building targets of 52,500 over the 20 years (2625 annually as opposed to the 2713 calculated above using the formula), so the formula didn’t have to be applied. For those interested an uplift of 7% was applied for second homes. When the next Local Plan is devised later this decade for 2030 onwards any formula will apply and Cornwall Council will have to adhere to it.
What may startle people is that the population forecasts and therefore the household formation projections from 2016 that were rejected by the government actually showed an increase in Cornwall’s population due to internal migration. So it would’ve led to more houses being built in the county- not less. Those figures were before COVID, which promoted a further increase in internal migration on top of that already predicted. The pandemic essentially accelerated what was an already existing trend. So any future housing need assessment has the potential to be considerably larger than we have now.
The death of the standard formula?
There are many critics of the formula with some saying it often inflates the need for housing in the least affordable areas which enables developers to build more houses in the areas that can produce the highest returns. This can interfere with other policies; for example the Levelling Up agenda in areas where affordability ratios are lower, so the formula would result in less uplifts and less homes being built. Absurdly, in some areas, using the formula would’ve meant stopping and decreasing house building targets where new homes were very clearly needed to aid economic growth. What it doesn’t take into account are the reasons why affordability is low or high in specific areas; this could hold the key to what housing should be built, where and for whom. There are others who agree with a central formula or mandate but don’t agree with the formula itself or the way its been developed or executed.
I do not believe authors of the original reports stating the 282,000- 340,000 figures, or subsequent others the government has relied on or referenced, ever intended the figure to be used as a general annual target, or a pre-determined outcome of an ill conceived equation, without any further enforceable intervention. It is a number that has been taken out of context from the reports and evidence in which it appears.
What the government left out when hammering their gigantic ‘not seen since the war’ building targets to the wall is that in both the Barker Report and the NHF and Crisis report those headline figures were accompanied by some clear instructions on what houses needed to be built to achieve some clear policy outcomes. In the 2004 Barker report it wasn't just about need but also about keeping house prices down to a more affordable level. The building targets weren’t met and the average real price increase has been 3.1% year on year since- nearly triple what the report said was the optimal outcome.
The pattern of events is the same with the NHF/ Crisis report; the 340,000 figure put forward, if you were to carry on reading, is made up of annual targets of 145,000 affordable homes – comprising 90,000 homes for social rent, 30,000 for intermediate rent, and 25,000 for shared ownership.
If 300,000 a year is calculated looking at housing needs across social, affordable and normal market housing, then the ratios of build types should be reflected in the target. 145,000 affordable homes is 42.6% of the annual development figure of 340,000. But that’s hard for local authorities to enforce or consequentially achieve when the National Planning Policy Framework (NPPF) states that where major housing development is proposed, planning policies and decisions “should expect at least 10% of the homes to be available for affordable home ownership.” So in that sense the planning policy could be seen as out of step with the development policy by around 32%.
Rishi Sunak’s Autumn budget confirmed a ‘£24 billion multi-year settlement for housing’, stating the government’s intention to ‘unlock’ more than a million new homes between now and 2026 (although I am unclear what ‘unlock’ means- if he could’ve said ‘build’ he would’ve). However whilst £11.5bn of the £24bn (previously announced so not new money) would be used to build a target of 180,000 affordable homes over 5 years, less than 20% of the overall target, only 32,000 of them will be social rented housing, which is a third of what is needed. So the same arguments concerning the mismatch of getting money to the correct housing market continues. This is despite the Local Government Association (LGA) predicting that council waiting lists could increase to 2.1 million by next year.
We have yet to see how Michael Gove MP, who replaced Robert Jenrick MP as the minister responsible for planning and housing in September’s reshuffle, is going to shake things up when the new planning bill finally appears. However it is now becoming clearer that building 300,000 homes a year is a tall order. Building 300,000 of the right homes and getting them to the people that need them is impossible. Since the formula was created in 2017 the government has delivered an average of around 230,000 homes a year. Cornwall has successfully added around 25,000 houses to its stock since 2010 and is on course to deliver its target of 52,500 by 2030. However the same problem persists; in the local plan the total affordable housing need from 2010- 2030 is 30,910 but the affordable housing due to be delivered by 2030 is 17,836. So it looks like we actually plan to fall short of the affordable housing target whilst at the same time applying a 7% uplift for second homes.
The unravelling of the standard formula is a good illustration of how inadequate the approach to solving the housing crisis is. The illusion of the formula has been that tying it to an affordability ratio seems like it’s using some sort of data to alter the market. But the aim of keeping house price inflation at the 1.1% annual target in 2004 is now a pipe dream. The market in the UK is now so out of control that some estimate 20 years of adding 300,000 homes a year would reduce prices by only around 10%.
The same housing reports the government relies on have also often recommended other measures such a reform of council tax, making it more closely related to property prices, establishing a goal for market affordability and giving local authorities powers to charge more for second homes to improve efficiency of the use of stock. None of these have been made policy, with successive governments instead choosing to focus on big build numbers or tax payer funded subsidies to the market in the form of Help to Buy etc. The Government’s aim is not to stop house prices rising, but to encourage home ownership without a cost to existing owners.
Terrifyingly, I haven’t even mentioned climate change and we are nearly at the end (just the blog- not mankind quite yet). Climate change should be a major factor in any housing stock assessment and building plan. The carbon footprint of building more homes when we don’t make full use of the stock we have is a rising drum beat that will get louder in the future. The fact that we are still building homes now that will need to be retrofitted in the future is another illustration of how the government isn’t taking it seriously. It’s the homeowner equivalent of having to constantly replace the ink in a printer.
The most depressing factor is how the formula is set up to fail society’s needs; the very people that provide the numbers for the house-building target aren’t considered in it’s scope or delivery. The number 300,000 was always the desired outcome and they worked backwards to invent a warped version of reality. It wouldn’t have worked so well for NASA if they had had the same approach.
It turns out the only rocket science needed to solve the housing crisis is how to get government ministers to start making the policy changes that society really needs.
Notes:
A few eagle eyed readers will have spotted I created the Cornwall 20 year calculation for 2010-2030 using data from 2014. I know this is statistically impossible (2014 is after 2010) but I needed to be able to make a direct comparison between Cornwall’s Local Plan (running from 2010-2030 but not adopted until 2016) to what would’ve been the outcomes had the formula been used.
I will create a separate page containing copies of or links to various housing reports going back to 2004.
Thanks to Dominika Roseclay from Pexels for the headline photo.
Comments