The housebuilding sector – then there were five?

It’s been happening for a while now but it looks like 2023 could be a significant turning point in terms of the structure of the British housebuilding industry.

Since the merger of Taylor Woodrow and George Wimpey in 2007, Taylor Wimpey, Barratt and Persimmon have been (figuratively) head and shoulders above the other listed housebuilders in terms of completion levels.   For years, they each completed more than double their nearest competitor (Bellway) and been the only companies building over 10,000 dwellings a year.

It looks like we’re now moving towards a Top Five rather than a Top Three.  Following the deterioration in market conditions associated with rising interest rates, the large listed housebuilders have responded quite differently in terms of their output.

Barratt have maintained their output between 2021/2022 (17,908 completions) and 2022/2023 (17,206 completions).  Similarly, Bellway’s (10,945 completions in 2022/23 compared to 11,198 in 2021/22)  and Redrow’s (5,436 completions in 2022/23 compared to 5,715 in 2021/22) outputs have also been relatively stable.

Persimmon and Taylor Wimpey haven’t yet released their annual reports yet. But the big story seems to be one of downsizing. Taylor Wimpey expect their completions to be about 10,000-10,500 this financial year.  Last year (2021/2), it was 14,154.

Persimmon have made a similar strategic shift and are expecting to complete even fewer.  In the half year update, it’s pretty clear that Persimmon’s output has contracted quite dramatically in response to changing market conditions and a strategic shift towards increasing the quality of their new homes.  They expect to complete around 9,000 new dwellings in 2023.  Last year, it was 14,868.  Compared to the comparable period in 2022, their completions had fallen by 36%. Accordingly, Persimmon’s gross profit had fallen by 54%.  Their gross profit margin had fallen very significantly from 31% to 21.5% with spending on land going down from around £416 million to just over £240 million.

As Bellway look to overtake Persimmon and Taylor Wimpey in trems of completions, Vistry may also emerge as the second largest housebuilder in terms of output.  Integrating Countryside Partnerships, they’re on course to complete more than 14,000 new dwellings. Compared to the first half of 2022, their output in the first half of 2023 had gone up from 5,409 to 7,143 in the comparable six month period.

Explanations for the divergence in strategies regarding completions are very welcome…

Landbanking: update on update redux

The CMA have produced a couple of working papers as part of their ongoing investigation market study of the housebuilding sector. I’m particularly interested in their paper on landbanking

They provide a fairly useful summary of the state of knowledge on landbanking and have gathered a lot of evidence.

Using our information-gathering powers, we have gathered data on the size and exact location of more than 5,800 individual sites held in the land banks of 11 of the largest housebuilders, giving us a more detailed overview of land banks across Great Britain than is possible from public data alone. To examine planning permissions at the local level, we have used commercially available data from the provider Glenigan.

Their estimates seem pretty good to me.

Our analysis suggests that the 11 housebuilders together own or control land equivalent to c.1.17m plots across England, Scotland and Wales. The quantum of land in long-term land banks is equivalent to c.658,000 plots, while the short-term land bank is smaller, at c.522,000 plots.

However, they seem to struggle to draw any firm conclusions based on this data.  On the crucial question of the appropriate size of land banks for operational purposes, they state that…

We are still considering how the size of these housebuilders’ land banks compare to what we might expect if their primary purpose is to allow housebuilders to manage their development pipelines.

Perhaps obviously…

There will necessarily be some lag between permission being granted and sites being built out, and this is likely to be greater for larger sites. In addition, not all land in short-term land banks is immediately ready to build out, given some will have only outline planning permission. Both of these things will imply a need for a land bank to some degree. However, we are yet to come to any conclusions on the overall scale of short-term land banks in GB, and what impact they might have on market functioning…

They identified about 10-15 local authorities where there are concerns that there is too much concentration of land ownership and/or control by a small number of firms but are admirably tentative about inferring too much from these findings.  The local authority boundaries may not capture housing submarkets…There may be particular local circumstances that have led to local concentration. So,

We have not yet reached a view as to whether we have concerns as to how the market is functioning in these areas – this is something we will continue to explore…At this stage, our analysis has focused on interrogating the datasets described above and some limited desk research. As such, it may be the case that there are important local dynamics we have not taken into account in this analysis so far. Recognising that LPAs are a relatively crude approximation to local housing markets, we are considering how to identify and take into account constraints from the most relevant areas surrounding the LPAs in question. We also intend to conduct further evidence gathering from stakeholders active in these areas to understand the competitive position they see on the ground.

Sounds like we need to wait and see.

Inflation shock: getting real about house prices

In my ongoing work on the housebuilding sector, the CMA investigation etc, I decided to look at what’s been happening to prices in the last couple of decades. I’ve struggled to believe the numbers that I found initially!

I’ve been trying to understand the growth in the profitability of the major housebuilders since the GFC.  Turning first to their build costs, I went to data on construction cost inflation from BCIS’ Private Housing Construction Price Index (PHCPI).  Between 2005 and 2022, they suggest that build costs had risen by around 108% in total. That’s about 4.5% a year on average and doesn’t sound implausible.

I then went to the ONS data on the simple average prices of new dwellings.  I was surprised to find that the average price of a new dwelling was £212,000 in 2005 increasing to £334,000 in 2022.  In percentage terms, that’s an increase of about 55%. Over the same period, Consumer Price Inflation went up by about 50%. The low rate of house price growth was not expected – c2.5% a year!  Sounds ok for CPI but not for house price growth – although the average prices themselves sound ok…Hmmm.

I thought that I’d better triple-check whether it was plausible that house prices were just about matching inflation.  I then checked Nationwide’s house price index.  The produce a dataset of UK house prices adjusted for inflation.  It’s not new houses only and they use RPI instead of CPI to standardise prices in real terms.  The time series is a bit longer – so it also incorporates the relatively high inflation rates of 2022 and 2023. 

When adjusted for inflation, the average UK house price peaked in Q3 2007 at £334,634 (in today’s money, at the time it was £184,131).  As of Q3 2023, it is £260,181. This represents a real terms decrease of about 22% since 2007 (if using RPI to adjust). Real prices had recovered steadily to £306,235 in Q1 2022 after which there’s been a sharp deterioration.

I wanted to check whether I was missing something and did a quick Google search as a sanity check. I found this article that (coincidentally) was published yesterday by Andrew Oxlade in This is Money.  He ‘recycles’ a chart from a Savills publication that basically tells the same story. I need a lot of reassurance 🙂

It seems hard to reconcile this type of data with narratives of runaway house prices and affordability etc.  I then decided to have a look at the data on household disposable income over this period because I suspected that part of the problem could lie with declining real incomes rather than rising real house prices.

In 2005/6, the median household disposable income was £29,073.  In 2021/2, it was £32,349.  That’s total growth of 11% over the whole period or just over 0.5% per year. Could this be the main source of the house affordability crisis? 

Explaining low growth in disposable income is for an economist…There’s been lots of debate about the poor performance of the UK economy – poor productivity growth, austerity, etc. to explain this weak income growth.  I’ll stick to real estate.

Hopefully, this wasn’t all obvious.

RICS “Guidance”

For once I find myself knowing about latest RICS guidance (they don’t seem to like to call it guidance anymore) on valuation.

Neil Crosby is the ‘technical author’ for the RICS’ Practice Information on Discounted Cash Flow Valuation that was published yesterday. Enjoy!

I hadn’t realised that they’ve promoted or demoted Paul Batho’s Comparable Evidence in Real Estate Valuation from guidance to a Professional Standard. This seems to involve putting an additional cover page on the old document from 2019. But it’s probably a promotion.

Going back to normal service, I hadn’t realised that Nick French had done some research for TEGOVA in 2020 on the use of comparable evidence in property valuation.  Enjoy even more 🙂

WFH or RTO?

Marking the bankruptcy of Wework, Bryce Elder had a long piece in the FT yesterday that reported on recent reports from KPMG and Morgan Stanley on working from home.

There’s lots of good survey data covering the US, France and Germany as well as the UK and some of the findings are unexpected.  The youngest category of office workers, so-called Gen Z, tend to spend most days working from home and Boomers spend least time. I didn’t expect that – albeit it aligns with personal experience…

I’m a bit nervous about the internal consistency of the survey.  Oddly, the preferences are reversed when asked how much time they’d like to work from home.  Gen Z seem to be spending more time working from home than they would like to!

The other unexpected finding is that their employers seem to be increasing the size of their offices. Bryce comments wryly “Weird. Maybe the buildings just look bigger when there’s no one in them?”.

If reliable, it does suggest a notable difference from previous intentions.  I’d thought that the mantra had become “Greener. Smaller. Better.”

Ed Zitron, in a Business Insider piece, frames the WFH debates as a power struggle between workers and senior managers.  He stresses the lack of any robust evidence to show how WFH affects productivity – especially in firms that specialise in collecting huge amounts of data  about all of us.  

It’s reasonable to wonder why Amazon, a company that has data on hundreds of millions of people and their decisions, is struggling to come up with hard numbers to back up its dictatorial push back to the office. Perhaps the reason is that the data supporting Amazon and other companies’ RTO policies is threadbare, relying mostly on a few studies that use sample sets of questionable usefulness to back up their claims that remote work is less productive…But the weakness of the evidence won’t stop bosses from making these RTO mandates. They indicate a failure of imagination on the part of management and a refusal to do the work necessary to create a positive company culture.

Poor referencing? CMA Consultation update

The Competition and Markets Authority have just released the responses to their second round of consultation on the operation of the housebuilding market.  The CMA had identified two issues which they considered might merit a market investigation reference – landbanking and problems with the adoption of public roads and other amenities by local authorities. There were 20 responses including most of the listed housebuilding firms (Barratt, Bellway, Crest Nicholson, Persimmon, Redrow, Taylor Wimpey and Vistry) as well as the Home Builders Federation.

Sticking to landbanking, the response of the housebuilders was fairly predictable.  It’s pretty well summed up in Barratt’s response.

The land banks of large housebuilders are not acquired or used for anticompetitive purposes. They are a necessary part of a large housebuilder’s business model and, as the CMA recognises in its Update Report, they are driven in part by the nature of the planning regime.

As ever, in most of the responses of the housebuilders, the planning system gets it in the neck, the knee, the eye, the back etc.

I had rather thought that the planning system was being used as a scapegoat by the housebuilders but then I remembered the reasonable housebuilders that I’ve met who can get genuinely emotional (mainly anger and frustration) about the uncertainties, vagaries, costs and delays that characterise the planning system.  I don’t think that it is performative.

A planning consultant that I know well constantly reinforces these exasperations. I have to remind myself that we didn’t design the planning system to be this way deliberately.  Rather that it is more like a bureaucratic version of a Heath Robinson contraption formed from lots of legacy policies with botched quick fixes attempting to reconcile lots of conflicting social and political forces.

It’s important not to be too parochial. These conflicting social and political forces and frustrating land use regulatory systems (perhaps simplistically but increasingly NIMBY v YIMBY/YIYBY) are happening in plenty of other cities globally.

The frustrating thing about the housebuilder submissions is that they are largely evidence free. Whilst they repeat the mantra that there are sound operational reasons for housebuilders to secure access to land for future development (Taylor Wimpey), they don’t actually say how much land they need to hold for operational reasons.

Swimming against the tide a bit compared to other academic commenters,  I’m almost embarrassed to write that the more I’ve looked at the topic, the more sympathy I have for the large listed housebuilders as far as landbanking accusations are concerned. 

There’s a number of reasons for this that I’ll try to set out briefly.

Firstly, land-banking ties up a lot of capital and creates risks to the security of this capital for housebuilders.

Secondly, there is a risk of confusing effect and cause here.  If housebuilders buy land, it becomes part of their inventory and it could perform well as an asset and it could prevent competition from other developers in the local market. It could do all three.  Is it fair to presume that they’re mainly hoarding? 

Thirdly, a proportion  of their short-term land bank plots with planning consent can’t be built out because they are on sites that need detailed planning consent and/or to satisfy other planning requirements and/or require on and offsite site works before works can commence.  The housebuilders could have said what the proportion was in their responses. 

Fourthly, a proportion of landbanked plots will be on sites that are being built out.  In simple terms, if you’ve just completed 200 units on a site in a single year that has permission for 1,000 units and expect to keep building at this rate, then you have a land bank of four years on that site.   As the average site becomes larger, it’s likely that this will be more of an issue.  Again, the housebuilders could say how much of their landbank consists of this category of plots in their consultation responses. 

Fifthly, in the US, the large housebuilders have shifted to a ‘land-light’ strategy and have been able to control rather than have to buy plots with a significant lag before construction.  The interesting thing has been that, when trying to minimise their exposure to investment in land, the US housebuilders still control around five plots for every one they complete – and still tend to own one or two as well. I should find out why this is. However, the key point is that, even when US residential developers are able to structure their land acquisition to enable them to buy plots a short time before construction, they still seem to need to control quite a lot of plots presumably for operational reasons.

Sixthly, it’s not just the US where the housebuilders have significant landbanks. The listed housebuilders in Australia and Ireland have larger landbanks in terms of years of inventory.

Finally, most of the volume housebuilders strategic land is controlled through options and these sites haven’t got planning consent.  Since the housebuilders typically haven’t paid for it yet, they won’t want the land value to go up (although they’ll probably be happy because the causes of growth in land values will usually be good for their business).  They typically want to get a consent on optioned sites. Otherwise they’ll have wasted a lot of time and money.  The sooner that they can get consent, buy the land and build out the scheme, the sooner they can recover their investment in land promotion and acquisition.

One more…there must be some operational issues as well.  Getting the right site in the right place at the right time is hard.  In terms of providing continuity of employment and workflow to contractors, there are likely to be challenges in ensuring a suitable site is ready for the workforce to move onto once a scheme is built out. It can take a long time and there’s lots of uncertainties in finding suitable sites, winning the bidding process and executing the transaction.  Alternatively, if prices and demand are falling, then they’re stuck with a site that may no longer be viable.

And another one…with the exception of Berkeley Homes, the housebuilders’ short-term land banks are all fairly similar in terms of years’ supply.  They all tend to hover between four and six years’ worth of plots. When the completions go down, their landbanks go down.

Too much sympathy and understanding?  I could go on about the fact that land values haven’t grown very much in the last decade, that the housebuilders don’t promote most of the land through the planning system, that SME housebuilders don’t really compete with the large listed firms…

I was surprised that they didn’t make a stronger, more evidence-based case in their submissions to the CMA. It’s not just, mainly or even the planners’ fault that they need five or six plots in their landbanks for every one they complete?  But that’s the main message from their submissions that seem to have been co-ordinated by Slaughter and May incidentally.

Oliver wants more – from UK housebuilders

In a week when the BBC have also been asking hard questions about the roots of “Britain’s housing crisis” (the wonderful Lucy Mangan reviews it here),  Oliver Wainwright today provides a pretty damning critique of the quality of the product delivered by UK housebuilders.  It’s a timely Guardian piece that reiterates a lot of regularly asserted (as Oliver acknowledges) allegations about the UK housebuilders over several decades.  They have too much market power, they build houses poorly, they make abnormal profits, they struggle to get skilled labour, enforcement of regulation is weak, their large landbanks reduce competition…

…a picture emerges of a system that is broken at every level. From the virtual monopoly of the few big housebuilders, and their iron grip over the limited supply of land, to the combative, buck-passing nature of procurement – laid out with such tragic clarity during the Grenfell Tower inquiry – to the lack of proper supervision on site, to the rapidly declining skilled workforce, a perfect storm has been created for a crisis of crap homes. The problems are not new. For decades, independent government reviews and all-party parliamentary groups have lamented the state of housebuilding and the focus on profit over product.

It’s quite a blame game with everybody blaming everybody else – the planners, building control, the housebuilders, the contractors, central government etc.  A lot of stakeholders want more – better quality, lower costs/prices, more supply, more choice, more consumer protection, lower environmental impacts etc.   Of course, the same debate and arguments are taking place in lots of global cities.

Personally, I would have liked to see a bit more systematic evidence of poor build quality rather than – admittedly pretty horrendous – case studies of poor build quality.  Whilst the guys from New Homes Quality Control feature heavily, the New Homes Quality Board (and its code) doesn’t feature at all. 

There was one obvious voice missing from the article – the housebuilders themselves.  I’m no expert on the typical quality of new build by volume housebuilders. However, I see in their annual reports lots of references to levels of customer satisfaction as well as profit margins, landbanks, sustainability etc.   

It’s also a bit odd that there is no reference to the latest Competition and Markets Authority’s ongoing investigation of the housebuilding sector.  The consultation exercise has attracted a lot of good analysis. Even the Homebuilders Federation acknowledge problems with build quality in their response to the consultation exercise

This phenomenon can be observed in the overall decline in customer satisfaction ratings for housebuilders overall during the period of rapid housing supply growth in the middle of the last decade. While the vast majority of buyers were still very satisfied with the build quality and customer service they received, the decline in scores for the sector as a whole led to a wholesale review of the checks and balances, the gaps in consumer protection and, ultimately, to the creation of the New Homes Quality Board.

I don’t think that I have read anything (produced independently) that has a good word to say about the housebuilders.  Surely, especially after the leasehold scandal, the consensus can’t be wrong?  I recently wrote an article on the land banks of the volume housebuilders.  It was titled “Housebuilder landbanks: inventory, asset or hoard?”.  The more I’ve looked into the topic, the more I wish I’d added scapegoat to the title.  But that’s still work in progress and I remain, as ever, uncertain and unsure.

A tasty Curry dish

I recurrently marvel and despair at the gaps in my knowledge…Given that a lot of the research that I do is about policy and market change, I really should know a lot more about theories of change and transition…

I was very interested to read Andrew Curry’s substack post about various models of transition and change on Saturday.  It’s noteworthy that the various models discussed to conceptulise sustainable transitions include problems of ‘lock-in’ and ‘backlash’ which have become quite topical.

I also marvelled/despaired to read about power and interest grids, actor analysis, stakeholder mapping (I had heard of that!).  In just a short read, I could see lots of potential approaches to analysing of lot of ongoing research issues.

Definitely worth the small effort involved in reading.  I certainly should be transitioning a bit more in an intellectual sense 😊       

Academic misconduct (HE not RE)

I read a couple of thought-provoking articles last week on higher education and the integrity of academic research more broadly.

The first was a very interesting case in the New Yorker on a case of apparent academic misconduct or problems of research integrity at Harvard and possibly much more broadly.  It’s a great story that’s hard to put down.  It looks at apparent data manipulation by leading behavioural researchers.  My worry is that it’ll give comfort to the stereotypical internet conspiracy theorists who claim that any evidence produced contradicting their twisted theories involves malevolent machinations by academics.  However, I think that, in the end, sunlight is nearly always the best disinfectant.

I sent a link to Brian Rosenberg’s book to a friend of mine who was in a senior leadership position in higher education. You can get a taste of it here.  The book reflects widely held stereotypes about the resistance to change that can characterise the cultures of academic institutions. In Tony Blair’s parlance, my friend has the scars on his back.

However, I then read Tim Burke’s review of the book which provided a robust response.  Having been in various institutions where there always seems to be someone trying to make a name for themselves by top-down ‘restructuring’ of something or other, it feels like we have nothing but change (whatever that means…we’re always making minor and continuous changes). I agreed with his take more than Brian Rosenberg’s.           

Of course, both can be right.  Academics can be obstructive, self-interested and, when empowered, able to prevent useful initiatives. I’ve seen it myself – usually in elite institutions.  Mainly, in my experience, we (academics in non-elite) institutions tend to be very interested in making improvements that can help our students and colleagues.  Increasingly, we’re disempowered and have never been able to prevent initiatives from ‘external’ managers which rarely never provide any explicit consideration of costs and benefits nor sometimes even a rationale. That reminds me – we’re due a full-scale rebranding at Reading as we replace our terms with semesters and restructure all our courses to fit the new semester model 🙂 It could be a great use of scarce resources – you just never know. It’s never evaluated and we’ll have already moved on to the next restructuring anyway.

German hushing

As ‘greenhushing’ takes off (I hadn’t expected this one), it’s good depressing to know that the UK government isn’t alone in its retreat from improving the environmental regulation of new and existing buildings.

The only surprise about the announcement that the proposal to introduce stricter MEES in the PRS sector has been abandoned is that some people seem surprised.  The government consulted on this proposal in 2020 and seem to have maintained radio silence ever since.  A cynic would argue that the recent announcement just confirms what we already knew. The National Residential Landlords Association seem worried that Labour will persist with the policy.

According to Gino Spocchia in the Architects Journal

Energy efficiency measures such as the Future Homes Standard, which applies to new homes, are still expected to come into force from 2025 under current proposals.

I have my doubts. However, a lot of industry and professional associations were pretty scathing about the policy shift largely on the grounds of not being able to make any plans when there is no certainty about policy.

Anyway, it’s not just the UK government.  In light of a sharp market downturn and other political pressures, the German government is also rowing back on similar policies. Guy Chazan reports in the FT

Germany has decided to shelve stricter energy efficiency rules for new buildings, in the latest example of a volte-face on environmental policy by a government scrambling to kick-start its weakening economy…Germany’s ruling coalition had originally agreed to impose energy efficiency rules dictating that new houses should consume just 40 per cent of the energy of a regular new-build. But the plans have angered the construction industry, which is already struggling with high interest rates, expensive materials and excessive bureaucracy. Economy minister Robert Habeck yesterday announced the new standard would not be introduced in this parliamentary term, which ends in 2025. He said a recently adopted law encouraging people to switch from oil and gas boilers to heat pumps would ensure climate-friendly heating systems in buildings from 2024. “Therefore I no longer consider it necessary to rush to introduce the new . . . standard,” he said in a statement. “We can wait with that,” he said, adding it made no sense to implement the law before European rules on buildings were finalised.