Did you hold out much hope that the Government’s long-awaited Planning Practice Guidance for Viability would finally clear the thickets, bringing both light and certainty in its wake? Me neither.
The problem of development viability – at least from a planning point of view – is, at root, a question of land value. How much of the increase in land value arising from the grant of planning consent should flow back to the land owner and how much can be captured by the Planning Authority to pay for affordable housing and infrastructure? On that crucial question – the question that the Guidance was designed to answer, the current draft leaves almost all of the muddles every bit as muddled as before. Its primary contribution is a layer of semantic confusion, which might have been designed to compel anyone who thought that they half-understood the question, to give up.
But before I dive into the weeds, I should acknowledge that there are a couple of bright spots. It is good, for example, that the Guidance finally gives practitioners a solid steer on allowable profit margins. It’s also sensible to suggest that developers submitting viability assessments to accompany planning applications will want to refer back to the assumptions the Council made when setting its policies. There is support for the use of benchmark data on costs. Finally, there is confirmation that all viability assessments should be made available to the public. None of that should have come as much of a surprise to anyone but it’s good to see it written down.
The problem, as I stated above, is the approach to land value. Let me summarise it as briefly as I can.
The Guidance seeks to resolve the tension between the Harman/Mayor of London Approach to Land Value, known as EUV+, and the RICS’s Market Value approach. The new approach is called EUV+ but the concept of Existing Use Value now subsumes Alternative Use Values. However, the premium (the Plus bit of EUV+) is now calculated by reference to the RICS’s Market Values rather than a percentage of EUV. The problem is not only that this is confusing, since EUV+ may now mean totally different things in different parts of the country, but it is also going to be fearfully difficult in practice because the problems associated with the Market Value approach have not been fixed.
Was that completely clear? No. I will try it again, a little more slowly. But this is going to be a long post.
The draft guidance is a hopeless fudge.
To understand why, it is worth going back a few steps and considering the role that viability plays in the planning system and why land value is so central to it.
For a site to be financially viable, it must generate a reasonable return to the developer (in the form of profit) and a return to the landowner (in the form of a land value). The developer’s profit was never very contentious and has now been put to rest. The landowner’s return must be more than the value of the land without planning permission – otherwise he won’t bring the land forward. The question is, (and always was) how much more?
There have been several attempts to answer this question, starting with the Harman report in 2012 and the RICS professional guidance, published the same year. The Harman report said that the starting point should be what came to be known as EUV+. That is, the Existing Use Value plus a percentage uplift – typically in the region of 20%.
The RICS positively pooh-poohed that approach. They pointed out that the only price at which a landowner could reasonably be expected to sell was the market value. The market value was, by definition, the value at which a buyer and a seller would transact in an arm’s length transaction, taking into account all of the circumstances of the site, the values achieved in similar transactions and all relevant planning policies.
Let me be entirely even-handed and say that both of these approaches were woefully incomplete.
The Harman approach is fine if the existing use value of your land is sufficiently high that 20% of it looks like a reasonable compensation for the hassle of promoting the site through the planning system. That simply isn’t the case for an awful lot of sites, especially greenfield land (would you sell a field for housing development at a profit of £5,000 per hectare?). The RICS approach would be sort of OK if there had been a lot of transactions for similar sites in the location (which there often aren’t) but the reason it really comes unstuck is because its logic is circular. If the market value takes into account the impact of all relevant planning policies, then how can you use the market value as a justification for providing less affordable housing than the adopted planning policy requires? You can’t.
There were attempts to fix that problem. The RICS was said to be on the verge of publishing updated professional guidance for years, but it never actually emerged. In the meantime, the Mayor of London (who doesn’t have to worry about greenfield land much) simply adopted the EUV+ 20% approach in London. Consultation on that closed four days before the publication of this new, national guidance. That’s why EUV+ may now mean different things inside and outside London.
So, how well does the new guidance fare at closing the gap? Well, confusingly, the approach calls itself EUV plus a premium. So it sounds like the Harman approach (or the Mayor of London’s approach if you prefer) but it isn’t. It is actually the RICS approach with a few additional bells and whistles.
The new approach starts from the Existing Use Value of the land. That is sensible – if the grant of planning permission results in a value that is below the Existing Use Value of the site then there is no reason to assume that the owner will release it for the new use. The development won’t go ahead. However, the Guidance now defines the Existing Use Value as including any alternative use for which the site has consent. That too makes sense – if you have planning consent for one use, you won’t bring the land forward for a new use that generates a lower value. My only quibble so far is that many practitioners used to treat this as a separate principle and referred to it by a different name, the Alternative Use Value, in order to avoid confusion.
What makes the new EUV+ different from the old EUV+ is the way in which the premium is defined. For Harman and Sadiq Khan, that was a simple percentage of the EUV. The new national guidance says:
“When undertaking any viability assessment, an appropriate minimum premium to the landowner can be established by looking at data from comparable sites of the same site type that have recently been granted planning consent in accordance with relevant policies.
The EUV of those comparable sites should then be established. The price paid for those comparable sites should then be established, having regard to outliers in market transactions, the quality of land, expectations of local landowners and different site scales. This evidence of the price paid on top of existing use value should then be used to inform a judgement on an appropriate minimum premium to the landowner.”
So, what you to establish the benchmark land value for your site is you work out what it is worth in its current use.
You then look around for sites which were previously in the same use but which have obtained the same sort of residential consent that you are seeking and which provided the full complement of planning obligations sought by policy. You will need to review enough transactions to work out whether any are anomalous. You then work out what those sites were worth before they received planning consent – although there is unlikely to be a transaction on that basis so you may need to use a benchmark. That allows you to work out the premium that they achieved, which you can now apply to the EUV of your own site and now you have a Benchmark Land Value that you can stick into your appraisal.
Now, I’m not saying that this is unworkable. It isn’t impossible to do in theory. What I am saying is that it will be hellishly difficult to agree on in practice.
Here are just four of the problems that practitioners will have to weigh:
- Land markets in some parts of the country are actually pretty thin. The number of transactions is finite. And development land isn’t a commodity – with one bit much the same as another – the value at which land transacts depends upon an awful lot of factors such as; its planning status at the time of the transaction, house prices in that location, the planning policies in place in that location and time, any contamination, the density and typology of the homes proposed, the continued viability of the original use. In order for a site to be truly comparable, all of these factors need to line up or, alternatively, you need a lot of transactions so that you can work out the average or you need to be making some heroic assumptions. For some use types, that may be possible but not for all. In Oxfordshire for example, you will doubtless have lots of greenfield transactions to review, and you may be able to form a view. But what about the conversion of an old nursery school to housing. How many local transactions of that type have there been in the neighbourhood recently? Or a former petrol station? In general, brownfield sites vary vastly more than greenfield ones. That is a problem because brownfield sites are, simultaneously, the ones which most often run into viability problems and the ones to which it will be hardest to apply this approach.
- Even if you can find good comparators to guide you, it isn’t clear whether the premium you have identified as the going rate should be seen as a percentage, a lump sum for the site or a lump sum per hectare. Each of these three approaches would give a different benchmark land value. For example, I have a petrol station of 0.3h with an EUV of £500k. I have found a direct comparator transaction which was also a petrol station of 0.25ha. Its value was £450k but, following the receipt of residential planning permission, it sold for £1.1m. If I apply the same percentage uplift to my site, my benchmark is £1,222,000. If I treat it as a lump sum then it’s only £1,150,000. However, if I do it as a per hectare rate it would be £1,280,000. Those three interpretations give me £180,000 of uncertainty and the guidance won’t help me resolve it. That calculation would get even more complicated if the two sites received consent for different numbers of units. And more complex still if it was two years ago and property values have gone up by 10%. What about if the lease on the old station was up but the one on mine isn’t? Finding more sites might actually make things even worse.
- Third, the new approach may put LPAs at a structural disadvantage. Viability assessments will be carried out on the basis of generic cost data – the new guidance specifically supports the use of BCIS data and the fact that the reports will need to be published implies it. But when land owners sell sites, they sell to the highest bidder. In general, the highest bidder will be the one who has the lowest cost base. Therefore, the prices paid for land in the open market may well be based upon costs below the averages in the BCIS index. If you do a viability assessment based upon generic costs and you use a land value benchmark based upon prices paid in the actual land market, you may not be comparing apples with apples. the cost ases will be different and planning authorities are likely to come off worst.
- Finally, and I’m sure that this a kink which will be ironed out through the consultation, because Alternative use Value has now been subsumed into the definition of Existing Use Value it currently appears that the same premium should be applied on top of AUV as well as what we will now have to call “old-skool” EUV. That is plainly wrong.
What this guidance represents is the triumph of the land market over the planning system. Under this approach, the viability of this site is always contingent upon the history of every other site and the story that one can tell to link them all together. It represents a defeat for the view that planning should control the land market than the other way around. I don’t say whether you should consider that a good thing or a bad thing. It’s just a thing.
It is probably true that no one approach could have solved the conundrum the guidance set out to solve but I think that a group of other approaches, used in concert, could have done. In my view, in order to be viable, any site would have to meet all of the following tests.
- Does it achieve a de minimis value? (Say £250,000/ha);
- Does the land owner receive at least half of the uplift in land value created by the grant of consent? (the so-called Shinfield test);
- Does the uplift in value represent a premium of at least 20% over the EUV;
- Is the land value achieved by this use greater than that achieved by any other use with an extant consent (no premium).
All of these tests are free standing – they can be assessed without reference to the other factors in the land market, they are demonstrably associated with uplift in land value (the premium over and above EUV) and they are clear. Many practitioners already use them as a common sense rule of thumb to check the outcome of more “sophisticated” methods. My guess is that, whatever form the official guidance takes, many will continue to do so – on the quiet.