Don’t Shoot the Messenger

We received a wide range of coverage when we reported at the beginning of February that landlord confidence was now worse than during the financial crash.

We also received quite a lot of interest, questions, comments and, in some cases, criticism about the forecast that some 500k properties would be sold over the next year or so, and that the PRS would shrink by up to 135k properties in real terms by 2021.

It is difficult to forecast what any market will do – witness the track-record of economists since the dawn of time – especially one as complex as the PRS.  As such, every prediction should be taken with a pinch of salt, but in order to explain why we believe our assessment is robust enough for public consumption, perhaps an insight into our methodology would be useful.

However, before we do the boring science bit we’ll confront some of the criticism upfront. A few are below:

Did we really expect our figures will change the Chancellors mind on this issue? Nope.

Did we expect sympathy from the media or wider public? No, but we did want to communicate that those tenants who are unable to buy, and who genuinely need to rent, will be the worst affected by the Chancellor’s decisions, despite him dressing up the changes as a good thing for tenants and first-time buyers.

Do we think landlords will follow through with their intentions? Some will. The rest? We don’t know. We have been very clear to caveat that.

Now to the science bit…

Predictions are only as strong as the foundations upon which they are built, so we start with the best data we can find. This means that we select our source material very carefully. In the case of ‘landlord confidence’ this is pretty straightforward.

For almost a decade we have surveyed landlords about their expectations in relation to various measures: capital gains, yields, the UK financial market, the UK PRS and their own business. Each respondent can rate the prospects of each metric on a scale of ‘very poor’ to ‘very good’.

We use this to calculate ‘optimism’ by subtracting those who rate prospects for their own business as ‘poor’ or ‘very poor’ from those who rate it as ‘good’ or ‘very good’.

I.e.

Optimism = (Very Good + Good) – (Poor + Very Poor).

graph1

Over the years this has given us the following, very illustrative, trend line, which shows confidence almost ten points lower than at the depths of the crash

The more controversial calculation, according to some commentators, seems to have been our assessment that around 500,000 units of rented accommodation will be sold in the near future.

It should be noted that this is an assessment based on what our members and landlords more generally tell us they plan to do, which is not always a straightforward and reliable indication of the actions individuals will actually carry out. However, the response appears consistent across multiple waves of data collection and we have attempted to use as conservative and robust a set of data and assumptions as possible.

It is not as immediately apparent as the optimism metric, but we are confident in the numbers and method used to generate our forecast, which I will try to elaborate on below.

So here goes……

We started with a forecast based on the Government’s own ‘UK Household Projections’ 1961-2037.

We supplemented that with information from the DCLG Private Landlords Survey concerning the typical number of properties owned by landlords, and the way in which they form their portfolios, in order to construct a reasonably robust picture of the sector and growth trends.

Next up we surveyed our members (twice) about their reaction to the Chancellor’s Budget and other announcements as part of two regular quarterly panel surveys. This allowed us to understand landlords’ stated intentions, and gave us the ability to cross-cut those responses with other information collected about the same (anonymous) individuals and companies.

As such we were able to draw out data about different types of landlord and what they state as their future intentions. For example, we were able to ascertain the likelihood of large portfolio landlords to sell relative to smaller landlords, or new market entrants.

This gave us a great deal of useful information such as that represented below.

graph2

Once we extracted these cross tables we were able to apply the ‘typical’ set of intentions to the model of the sector constructed from accepted government data, and predict the number of transactions landlords say they are likely to enter into.

Erring on the side of caution we set out to produce a ‘conservative’ set of figures. Consequently, at every step we chose to use data at the lower end of the spectrum stated by landlords and assumed that all portfolios were likely to exist at the smaller end of the ranges, e.g. landlords who told us they would sell ‘some’ properties would be assumed to only sell one property in the majority of cases.  Likewise, landlords with a portfolio of more than 20 properties were assumed to own ‘only’ 20 properties, when in reality the number may be many more.

This led us to the forecast that around 500,000 properties will be sold by landlords – had we made more liberal assumptions the figure could have been much higher.

More caveats…

As you can see from the method above we actually understated the figures – for fear of being labelled as sensationalist – so the situation could be a lot worse if landlords actually sell in the numbers they say they will.

And that is perhaps the biggest caveat of all. We won’t know whether landlords will follow through with their intentions to sell, and these figures simply represent the current sentiment that exists. It doesn’t mean it’s what they will do, or what will actually happen.

How our story is reported is out of our control but we have a duty, as the largest representative body for landlords in the UK, to reflect landlords’ sentiment and to say it precisely how it is.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s