Land: Inventory or Asset?

With Sajid Javid almost haranguing the house builders about land-banking, perhaps it’s no surprise that the biggest of the lot of them (in terms of revenue anyway) commissioned a report on that very question from an economics consultancy.  My colleague, Ed Shepherd, pointed me towards it.  It’s a very interesting report too that provides a lot of good insights into the residential development process and a lot of good data on the participants in the residential land market.  I’m not going to go over the various allegations.  The claims of oligopolistic behaviour by the volume house builders have been well aired by Conservative minsters.

The implicit argument of Barratt’s consultants (Chamberlain Walker Economics btw) is that house building does not lend itself to lean production and, in the jargon of Just-In-Tine methods, organised flow scheduling throughput.

Ideally (they claim reasonably that) house builders want to have another shovel-ready plot ready when they have completed a dwelling.  Given it takes so long to make a plot shovel ready because of time taken to

  • Obtain planning permission
  • Discharge post-consent conditions
  • Implement lead-in, site preparation etc.

…and the need to take into account

  • Failure rates or site wastage (sites that go wrong)
  • Business growth of house builders
  • Difference in market absorption rates for large sites.

…As a result, they find that volume house builders need to have approximately four plots with planning permission (but still with undischarged conditions etc.) in order to have one that is shovel ready.  Conveniently, but (to be fair) plausibly, this is roughly the land inventory that the major housebuilders have.

The modelling demonstrates that a stock of 1.25 million planning permissions (1 million detailed-) would be needed for 250,000 home completions a year in the ‘zero growth’ steady state. This compares to a stock of around 0.8 million planning permissions (0.7 million detailed-) currently.  That’s a shortfall of around 450,000 planning permissions.

A lot of interesting further data is presented in the report on the role of non-builders.  Although I’m not convinced by their definition of non-builders

‘Nonbuilders’ are a very heterogeneous group and include landowners, land promoters, RSL’s, the public sector, operational and other businesses securing planning permission for other business reasons

RSLs should surely be defined as builders?  Anyway…

55% of all planning permissions in England are not held by builders at all. 87% of outline planning permissions are not held by builders.

There’s also some statements of, what seem to be, the obvious

…builders are more likely to have started their detailed planning permissions.

Maybe I’m missing something?

The elephant (calf?) in the room is also that the strategic land holdings of the housebuilders aren’t mentioned.  Personally, I don’t see buying land without planning permission or obtaining options on land without planning permission as land-banking but the house builders seem to be trying to have it both ways.  They can’t claim not to be land investors when it suits them and then, like Taylor Wimpey below, to say that land investment is a key part of their business model.

A key strength for Taylor Wimpey is our strategic pipeline. This land, which has no residential planning at the time we take a commercial interest, affords significant protection of future returns with a high embedded margin and, importantly, enhances our short term landbank when converted. We have the largest strategic pipeline in the sector which stood at c.108k potential plots as at 31 December 2016 (31 December 2015: c.107k potential plots). During 2016 we converted a further 9,519 plots from the strategic pipeline to the short term landbank (2015: 8,660 plots). With a significantly lower cost and greater control over the planning permissions we create, we continue to seek new opportunities and added a net 10.8k new potential plots to the strategic pipeline in 2016 (2015: 5.8k). In the year, a record 51% of our completions were sourced from the strategic pipeline (2015: 47%). 

But it’s the activities and motives of non-builders that we know little about.  Even companies like Legal and General Investment Management are getting into strategic land investment.

It’s not necessarily land-banking (depends on how it’s defined) bit it’s all pretty opaque and could usefully be ‘opened up’.  We don’t know much about the proportion of non-builders that are holding housing land with planning permission as an asset class (as a long-term store of wealth) rather than as a factor of (housing) production to be released when required.


Long-term stock selection?

Writing in the FT, Merryn Somerset-Webb steered me towards an interesting paper by Henrik Bessenbinder based in Arizona State University.  He takes a long term look at the performance of individual listed companies.  In contrast to a typical real estate asset, he finds that individual common stocks tend to have rather short lives. The median time that a stock is listed on the CRSP database between 1926 and 2015 is just over seven years.

However, the importance of a small group of companies to the overall performance of the stock market is really amazing.  Looking at the equity premium puzzle, he finds that most of the premium is due to the performance of a small sub-set of all companies.

I calculate that the approximately 26,000 stocks that have appeared in the CRSP database since 1926 are collectively responsible for lifetime shareholder wealth creation of nearly $32 trillion dollars, measured as of December 2015. However, the eighty six top-performing stocks, less than one third of one percent of the total, collectively account for over half of the wealth creation. The 1,000 top performing stocks, less than four percent of the total, account for all of the wealth creation. That is, the other ninety six percent of stocks that have appeared on CRSP collectively generated lifetime dollar returns that only match the one-month Treasury bill.  

Although, like Python’s parrot, it has long been pronounced deceased, Bessenbinder muses on the implications for CAPM.

The Capital Asset Pricing Model (CAPM) in particular implies that each individual stock’s expected return premium is the stock’s beta times the positive expected market-wide premium. Given positive betas, each stock’s expected return premium should be positive. Note, however, that implications of standard asset pricing models, including the CAPM, are with regard to stocks’ mean excess return, while the fact that the majority of common stock returns are less than the treasury rate reveals that the median excess return is negative.  

This type of work highlights to me how different equities seem to be compared to real estate assets.  There’s no comparable work out there (that I’m aware of) on long term holding period returns for real estate assets.  However, it’s hard to imagine such skewness and the prevalence of negative returns in individual buildings.  There isn’t the same “winner takes all” (or nearly all) potential in real estate markets.    Unlike successful companies, the key issue is that successful buildings are limited by basic physical constraints in the share of the market that they can capture .  A building just can’t make itself bigger in order o capture increases in demand.

Buildings can of course make negative returns.  Just look at Detroit for extreme examples.    We do see large variations in returns over the long-term even within sub-markets.  You could look at the holding period returns of residential compared to office investments in the West End over the last 40 years.   There’s a research project here somewhere…

Stable Brand?

Unlike many other products, it’s interesting how real estate developers tend to build buildings rather than build their own brands.  We tend to see it where services dominate e.g. for Regus and WeWork.  I’m sure that that the large real estate services firms such as CBRE, Savills etc. are fairly brand conscious.  There’s a very good overview of the volume housebuilding sector on the Brand New Homes website.  Hard to tell how trusted they are but the raft of recent stories of poor quality homes, leasehold mis-selling etc. won’t have done the image of the sector any good.

Buildings themselves can have a brand – often associated with a starchitect…Normal Foster and the Gherkin, Richard Rogers and the Lloyds Building, Renzo Piano and The Shard.  Individual shopping centres tend to be pseudo-branded – but I doubt that many consumers have heard of Intu, although many will have heard of Westfield.  I wonder how much of a premium Unibail-Rodamco paid for the Westfield brand compared to Hammerson’s value of the Intu brand?  Obviously brands have traditionally been really important to hotel chains.

With the exception of the soi-disant Stable Genius, very few developers create their own brand to market themselves to tenants etc.  I suspect that the closest that we have in the UK – and it isn’t close – is the Candy Bros.  There’s an interesting piece in the Guardian today about how the Stable One has been able to monetise his brand across many buildings that he doesn’t own.

Plastering the Trump brand – which can cost tens of millions of dollars to lease – on your luxury hotel or apartment complex once added a veneer of prestige and upped profitability; Trump used to boast it would increase a property’s value by 25%. Licensing his name certainly seems to have increased his personal fortune. A financial summary Trump issued when he kicked off his presidential campaign in 2015 valued his “real estate licensing deal, brand and branded developments” at $3.3bn – the most significant single source of what Trump then claimed to be an $8.7bn total net worth…While these numbers are impossible to verify and have been the subject of much debate, it is unambiguous that the Trump brand has traditionally been a source of considerable value. That may no longer be the case.

Maybe office landlords struggle to differentiate themselves in terms of providing a commodity.  I doubt that houses buyers want their homes with marques like their cars.  Should/could Barratt team up with trusted brands such as John Lewis or Miele and other prestigious  potential suppliers?  Hard to envisage.

Real active asset management options

I’ve just got around to reading the findings of the IPF-commissioned An Investigation of Hurdle Rates in the Real Estate Investment Process carried out by a team from Cambridge and Aberdeen.   Focussed on, er, hurdle rates, it provide a good overview of the challenges of applying mainstream financial theory to commercial real estate practice.  An added bonus is the good contextual material on the institutional investment processes.  Of course, they really should have referred to some outstanding previous work for the IPF that address similar topics 🙂

Different types of real estate asset were captured quite well.

Following Brennan and Trigeorgis (2000), there are effectively three types of investment projects. In the first, the investment manager is confronted with a project that has exogenous cashflows that cannot be altered by the manager in the future. In real estate, such a project would include some of what are called ‘core’ investments and would correspond to a fully rented building with long-term rental contracts. Corporate finance textbooks are very clear about the investment criteria that should be used for projects of this type. In the second type of investment project, future cashflows can be altered by the manager. In real estate, such types of project would include those that are called value-added or opportunistic, depending on the degree of flexibility that the project brings. A building with vacancies that can be filled or that has rental contracts that can be adjusted in the future would also be such a project. Another example would be a piece of land that the owner could develop in the future. The landowner knows today that she will wait for the best moment in the future and this makes the cashflows endogenous. Corporate finance theory prescribes that real option models provide the appropriate investment criteria in this situation. In the third type of project, strategic interactions between market participants play an important role. For instance, two developers might be interested in opening a new shopping centre, but only the first will be profitable.

This really chimes with something that I’ve tend to bang on about in our real estate asset management modules.  There’s a lot of variation among real estate assets in their potential for active asset management or adding value (alter the cash flows).   This is really due to variations among assets in building quality, intensity of site development, tenant numbers and qualities, (unexpired) lease terms, local market conditions, intensity of use, location etc.

There’s a strong relationship between asset quality/risk and active asset management/added value/real option opportunities.  High quality/low risk (core) assets tend to have little potential to add value by asset management (alter the cash flow).  Assuming that the core asset is fairly new, and high spec, tenants have strong covenants, best-in-class services are in place at competitive prices, weighted unexpired lease terms are long etc.  It is pretty difficult for even the most brilliant asset manager to add value by changing the building, the tenant, the leases, the service offer etc.

In contrast, higher risk/lower quality assets often require upgrades and tend to have potential to add value by increasing density (extending or reconfiguring the building), using the building more intensively, upgrading and updating the building and the infrastructure, improving the service offer through efficiencies and/or additional services, improving tenant quality and mix, improving lease quality etc.  In particular, where local market conditions are buoyant and leasing conditions are favourable, there can be significant rental and capital value uplifts from ‘re-positioning‘ assets.

Such real options tend to be easier to identify with hindsight and much harder to put a value on.

Hot air bubble?

There was an odd piece in the Observer yesterday on WeWork by technology writer Evgeny Mozorov.  The piece was basically trying to frame the Wework share price phenomenon as a paradigm “This time it’s different” shift in real estate.  I’m a sceptic.  When I see things like

 as a technology company, its main asset is its data, not its properties and its rapidly expanding size allows it to extract and analyse data related to their use and under-use (“buildings are giant computers”, says its blog). Armed with the data, it can then offer tenants flexibility on space, furniture and leasing.

Really?  Is flexibility adding that much value to the commodity of space and fittings?

WeWork is expanding in many directions. It has launched living spaces, where members can rent flats above their workplace. It has launched a wellness centre. It has acquired a coding school, where its future members might learn to code. It has announced an elementary school that will treat students as “natural entrepreneurs”, thus allowing their busy parents to see more of their kids – at work.

Feels like over-reach and reminds me of the hype.

“Our valuation and size today are much more based on our energy and spirituality than on a multiple of revenue,” its co-founder, Adam Neumann, told Forbes.

Anyone interesting in shorting it?

Energy consumption and house prices: from Stockholm to Dublin

There’s a recent paper on the relationship between energy consumption and house prices in Sweden published in Energy Economics that I missed last year.

Wahlström, M.H., 2016. Doing good but not that well? A dilemma for energy conserving homeowners. Energy Economics60, pp.197-205.

The title rather anticipates the main conclusion.  Some of the findings look odd and I think that I may be missing something in the results.  They’ve got a sample of around 77,000 house sales in Sweden that took place in 2009 and 2010.  Probably the only thing that you’ll remember is that there’s an average of 107 frost days.  The average heated floor area looks large at 167 sq metres with an average plot size of 1333 sq metres which is even larger than expected.   There’s quite a mix of heating systems – electricity only, district heating system, biofuels, various heat pumps and mixtures of these.  The measure of energy consumption seems a bit vague.  I’m not really clear whether it is actual consumption or modelled consumption.

…energy consumption is defined as the energy purchased for heating and cooling and for household purposes during the preceding twelve months and is expressed in kWh. The specified quantities purchased are usually based on a combination of information from the owner and/or calculations made by the EPC expert.     

Using standard econometric techniques – that admittedly I’m no expert on – they find an apparently unexpected positive relationship between energy consumption and house prices.  In terms of individual attributes, they find that some features such as a ground sourced heat pump produce an 18% price premium.  Given a median price of €217,000, with ground source heat pumps costing close to €20,000, then it’s not a crazy figure but, as acknowledged in the paper, it looks high.

The author does an interesting model where, although many of the energy features are statistically significant, they are omitted from the specification. Without any loss of explanatory power, the results of this model find no effect of energy consumption on price.  One possible interpretation of this finding is that the various energy features may be capturing the effects of potentially omitted variables.  For instance, it may be the case that dwellings with ground sourced heat pumps tend to have other positive attributes that are unobserved i.e. such dwellings may have superior heating systems and may also be superior in other ways that were not recorded.  Turning to the finding that there is a positive relationship between energy consumption and house price, could the explanation be as simple as more affluent house buyers buy more expensive properties and also consume more electricity?

Turning closer to home, there was another paper published in Energy Efficiency that I also missed last year.

Stanley, S., Lyons, R.C. and Lyons, S., 2016. The price effect of building energy ratings in the Dublin residential market. Energy Efficiency9(4), pp.875-885.

Drawing upon 2792 list prices from Dublin in the period 2009-2014, basically they try to improve on one of their previous studies – which they do.  Samuel Beckett might say that they’ve failed better.   Overall their results look plausible.  However, it’s interesting to note that they find that properties rated EPC B1 have a discount relative to properties rated EPC C1 and the coefficient for A3 properties is actually positive (but not statistically significant).   My suspicion is that properties with A3 or B1 ratings are rare in two ways – a conventional and an Irish way.  Firstly, there’s hardly any of them.  It’s not clear from the paper but I’d guess that there’s three B1 rated dwellings and one A3 rated dwelling.  They’re also rare in that they may be unusual types of building in other ways than energy efficiency and may not be as attractive to more many mainstream buyers.