Last week the housing ministry announced that Local Authorities must revert to using the 2014-based housing projections and ignore the 2016-based projections recently published by the Office for National Statistics. Their problem? The numbers were too low.
Local Authorities are required by law to form a Local Plan which determines among other things the number of homes the council intends to build over the next 15 to 20 years. This has been a difficult task over the last few years because the government have kept changing the rules. Nearly half of Local Authorities are still without an up-to-date plan as their assessed needs for housing have steadily risen sending them back to the drawing board each time they are ready to go forward.
In the 2017 general election the government pledged to get councils building one million homes by 2020 and another half million homes for the next two years after that. What they are actually doing now is forcing councils to plan for a total of 300,000 homes per year for the 20 year duration of their plans and beyond. A much more ambitious and totally unachievable aspiration.
In the past it would not have mattered too much if councils over-planned. The big developers could have tried to take advantage by cherry picking the sites that gave them the biggest profit but at least the planning authority would keep the power to control the process of approval. The council planning committees could demand that infrastructure was put in place with density and provision of affordable housing meeting their requirements. That is no longer the case. In July the government published a new National Planning Policy Framework just before the summer recess with no time for parliamentary debate. It includes a Housing Delivery Test that means if councils fail to deliver almost all their housing target then they will lose control of planning, leaving the larger developers to pick the sites they want with regard only to the minimal conditions for sustainability built into the NPPF. Never mind that it is the big developers who deliver most of the housing. The council can only enable development by releasing land into the hands of the larger builders. They do not have the capacity to build large numbers of homes themselves. The government were told this repeatedly but would not listen.
Last year the ONS published their 2016-based population projections which came out 19% lower than the previous 2014-based projections. The reasons for the drop were perfectly legitimate. Net migration was falling and increases in life-expectancy were not improving as fast as expected. If anything the numbers used are still too optimistic. At that time the government was consulting on their plan to introduce a standard formula for housing targets based on the ONS housing projections. These in turn depend on the population projections so it was already clear a year ago that the planed housing targets would be likely to fall with the update from 2014-based to 2016-based projections. The ministry plodded on.
So in September with the legislation in place the new 2016-based housing projections came out. Instead of being 19% lower as expected, they were 24% lower. The extra 5% reduction came about because of changes in methodology for projecting household formation. To get from population projections (the number of people) to household projections (the number of houses) you have to divide by the average number of people per household. The way the ONS do this calculation is complex but it comes down to a question of how many people are expected to occupy a house based on various demographic assumptions.
Between the census of 2001 and the census of 2011 the average number of people per household had dropped from 2.4 to 2.3 continuing a trend that had run for decades. In compiling the 2014-based projections, the ONS had assumed that this trend would continue down to about 2.2. A drop of 0.1 means that over a million new homes are needed in England so this is a major consideration.
By time the ONS came to produce the 2016-based household projections it has become clear that the average number of people per household had actually crept back up to 2.4. The method was adjusted accordingly allowing for a future number between 2.2 and 2.3, but higher than assumed before. This led to the extra 5% reduction in the housing targets and a lot of controversy.
The big question is this - are there more people per household because of changing lifestyles, or because not enough houses are being built? If people live longer there will be more people living as elderly couples or alone, bringing the numbers down, but the projections for longevity have fallen. Perhaps some would prefer to live together in the specialised accommodation that the government is pushing for. If people get divorced they will double up the number of houses but the divorce rate has fallen leaving the average household size higher (at least it is a good reason this time.) If young people leave home to live by themselves or start a family, again the household size goes down, but are people staying with their parents because they cannot afford a house, or is it because they go to university delaying the age at which they become financially independent? Maybe some are just enjoying the single lifestyle now that technology frees us from being a slave to home chores. Nobody really knows the answers to these questions because nobody has done a proper survey.
Whatever the answers they can only affect the 5% drop in household numbers due to changes in household formation rates. The 19% drop due to changes in population projections is another matter. It also has some uncertainty but mostly it is dependent on the unknown size of future net migration. The rate assumed in the population projections is still much higher than the "tens of thousands" that the government has promised to bring with immigration controls. If the government wants to claim that the numbers should be much higher than the projections, then they need to admit to their plans to increase net migration well above the current levels of about 250,000 per year. They can't have it both ways. Either it is low net migration and low housing projections or higher net migration and high housing targets. Which is it?
Instead of facing these questions the government has thrown back to the numbers from the 2014-based projections that we now know to be wrong. The targets are built into legislation. It is very hard for councils to avoid using them. Even the green belt constraint is no longer a good enough reason to avoid unpopular developments when the Local Plan goes to examination, although again the government will not admit it openly. Planning Authorities who already agreed a lower target will not get away with it for long either. Housing targets now have to be reviewed at least every five years, sooner if there is a significant unmet need in the plan verses the updated standard formula.
So councils will commit to building 300,000 homes per year but there may only be demand for about 160,000 homes a year according to the best projections. What will happen? The large developers don't build houses that are not needed so targets will not be met. Starting from next month, local authorities must calculate what percentage of their annual target they have delivered in November each year. When they fail to build enough houses the consequences are spelt out in the NPPF. No excuses will be accepted. It is no use the council pointing to an economic downturn. They can't say that they did not build because they did not have the funding for infrastructure. Even if the bigger developers just decide that the CIL is too high and they prefer to build in the next door borough instead, that will solely be the fault of the council. With unachievable targets forced on councils by the government it is absolutely inevitable that a large proportion of them are going to fail the Housing Delivery Test. There are no mitigating circumstances, no tribunal, no allowance for a judicial review. The consequences in the NPPF are automatic and unambiguous.
Some consequences of failure will be phased in over three years. However, paragraph 73 of the NPPF says that if the council delivered less than 85% of its target over the last three years it must add a 20% buffer to its target for future years. The remedy for a target that is too high is to make it higher, apparently. Where councils do not have an adopted Local Plan the target will be based on the standard formula, now using the higher 2016-based projections. This is going to affect a very large number of councils next month. Then from November 2019 onwards they will have to aim for a target 20% higher, and from November 2020 onwards they will be in trouble if they do not meet 75% of that target (i.e. 90% of the standard formula or the target in the Local Plan.) If nothing changes I predict that well over half of councils will not make the cut. Failure at that point triggers paragraph 11 part (d) of the NPPF whereby planning decisions will be based on the policies in the NPPF alone, and no longer on the policies of the Local Plan. The green belt constraint will still apply, but in areas released from the green belt the council will lose control of density targets or the amount of affordable housing that is to be included. The euphemism for this situation is "the presumption in favour of sustainable development". Unless the housing ministry executes a sharp U-turn, it is a phrase we will be hearing a lot in the future. For failing to deliver the houses that the council enabled by releasing land to them, the large developers will be rewarded with planning on a silver plated platter.