I’ve written before here at some length about how development appraisal techniques seem to have somehow become isolated from mainstream practice and theory in capital budgeting, project appraisal etc. Judging by what Argus Developer offer, appraisal textbooks, evidence from viability statements and anecdotes from market contacts, there’s a whole mix of practices going on out there. Many of them would be frowned upon by anyone from a mainstream finance or accountancy background.
In the past I don’t think that this has really mattered much from a practical point of view because many of the theoretically dodgy techniques and assumptions have been applied by developers (and their advisers) who can usually look after themselves. In terms of risk in the development process, a dodgy appraisal model and assumptions are just one of many risks and uncertainties that they accept when evaluating a development project – even if they don’t realise it. The limitations of the model and its assumptions are usually fairly minor compared to largely unavoidable risks of getting the inputs (projected revenues and costs) wrong. Put simply, a dodgy model doesn’t really matter when you’ve got lots of unavoidably dodgy inputs. However, it tends to matter more when games are played with affordable housing in development appraisals used to assess viability in the planning system.
In corporate finance, the basic tenets of applying discounted cash flow techniques to project appraisal are fairly straightforward. Estimate all the relevant cash flows and discount them at the target rate of return. If the NPV is zero or higher, then the project is viable. Alternatively, the NPV can be the worth of the investment opportunity – in a development context this is usually the land.
It’s clear from various pieces of project appraisal guidance that finance costs, contingencies and overheads are not relevant cash flows. Finance costs are implicit in the discount rate. The investment and financing decision should be separate. Contingency is really a risk management issue. It’s not an actual expected cash flow Overhead costs are largely fixed and taking on a project often has little incremental impact on them. However, contingency and finance costs (to a less extent overheads) are routinely included as costs in development viability appraisals in the planning system. An even bigger issues is the fact that viability guidance explicitly advises that expected changes in the cash flows should be ignored i.e. no forecasting. Ignoring something as basic as the time value of money, profit is also included as a cash margin.
So it’s a bit of a mess. It may also enable seemingly large returns to the developer to be ‘allowed for’ in a development appraisal. Let’s look at the cash flow of a highly simplified development project. (Just because it’s highly simplified doesn’t mean that the insights from it aren’t valid. It’s simplified in order to highlight the key issues and to avoid the – in this case – irrelevant detail). It’s the typical development DCF assuming interest at 2% per period, a 5% contingency and profit at 20% of GDV.
So – if £1,100 is paid for the land, a negligible amount is left over once normal development costs are taken into account, the developer gets a profit, a contingency is allowed for and interest on all costs is covered.
If we look at this cash flow using non-dodgy project appraisal theory, ignoring taxes it would look something like this.
It’s a bit simpler because it includes only relevant cash flows.
Is an IRR of 10.5% per period enough? Well, it depends on the period. We were allowing for £400 developer profit in the standard method above. When we take out the contingency and interest costs because they are not relevant cash flows, the projected profit increases – by another £100.
10.5% IRR per annum would typically be regarded as quite low for a development. 10.5% per quarter (about 49% per annum) would typically be regarded as quite high. Whether it’s quarterly or annual, the profit remains the same!
The simplification of taking profit as a cash margin (calculated as a % of total revenues or costs) for all schemes is really highlighted here. A £500 profit in one year is a lot better than a £500 profit in four years – but the typical development appraisal approach assumes that they are the same!
The difficult question is for another blog – how do we estimate what the target IRR should be for a development project?
There is also a bit of a paradox here. In a planning context, this analysis suggests that developers are including “irrelevant” costs because they are standard in the typical model. As a result, there’s less money for affordable housing and other planning obligations. Assuming that they’re using the same models to price development sites, the paradox is that, in a market context, it would suggest that developers would also systematically underpay for sites. This seems unlikely. It wouldn’t be possible if land markets were competitive – which they seem to be. It goes back to that difficult question of the appropriate target rate of return for a development project.