A Drunk Monkey Takes on the Experts in Price Predicting

There is an obligatory slogan after any investment product is advertised,

 "Past performance is no guide to how this will perform in the future".  

In this article I will try to claim this is completely wrong, as far as UK property is concerned.  I will assert that actually property that has done badly over the last 10 years, will do better than most in the next 10 years.  And vice-versa.  We will also look at other forecasts which are made by actual experts, and determine if they are more or less accurate in their predictions than a drunk monkey throwing darts at a dartboard.

The Art of Forecasting and Guesswork

There are so many different forecasters out there, many from huge and reputable firms, that offer guides to property values in the next 5 or 10 years.  Some forecasters are even paid by the government, ie our taxes.

 When you look to their forecasts over the last few years, and see how they did, you can often be left feeling like they are just throwing darts at a board and making it up as they go along.  Perhaps they are.

Take a look at the scene the predictions made for 2007-2008 in UK prices, in early 2007, just before the market crashed around UK:

- Capital Economics - increase of 3.5%
- Halifax Bank - 4% increase, with Northern Ireland being the driver
- Hometrack - 4% increase
- Nationwide Building Society - 5-8%, with London doing best.
- Council of Mortgage Lenders - 7%
- The Royal Institute of Chartered Surveyors - 7%
- Savils - a range from 7-15% across the UK
- Assetz - Increase 8-10%

Not only did they mis-judge the market completely, not one spotted the cliff which was just about to be dropped off, just a couple of months away and indeed drops of 20-30% were just around the corner.  That's a massive difference.  How could experts be so far out in their predictions?

And then after this "car crash of predictions" experts like Savils / Oxford Economics in 2012 gave it another go and claimed that prices in UK as an average would increase by 0.5% in 2013 and 1.5% in 2014, followed by 2-3% growth per annum.  Immediately after this report came out, double digit growth became the norm and up to near 20% annual increases now in London and Manchester for example.

Seems like they are throwing darts, and after a few beers at that.

Think I am making this up?  Just type "house price predictions - 2007" or any other year you care to mention into Google and see what was in the media at the time and compare it to reality.  There seems to be no correlation at all.

What's their latest punt?  Here's Savils' latest roll of the dice:

So, basically give or take it, everywhere goes up by 25% in 5 years in UK now according to them.  Nice and simple, you might even believe it, seeing as they have gone to the trouble of putting house shapes to the numbers as above, adding a degree of authority.

I would like to place a bet that this is as hopelessly inaccurate as all the rest of them.  Anyone want to get the other side of that one?  Please mail me or tweet over and we'll set that up for some fun.

Could a drunk monkey, armed with some darts in their hand, do any better?  Let's see.

Past Falls Indicate Where Future Gains Lie?

Right folks, this ain't going to be scientific.  Read the reports from the big boys like the above if you want the "science" and the seemingly daft predictions made from that.

I just make the point here from reading say the latest Nationwide report on prices which shows the prices over the last 10 years down to city level. 

Anyhow, the latest report is here:

http://www.nationwide.co.uk/~/media/MainSite/documents/about/house-price-index/Q1_2014.pdf

What is interesting here are the combination of the thoughts of the approx 20 year UK property cycle [see my previous post on this], the 10 year view the reports above give and the view from most onlookers to the UK property market on the rough 10 year average in price doubling time we experience in the UK.

So, how has it gone over the last 10 years?

City                                                                

London                83% [average of all Boroughs]

Edinburgh           28%

Cardiff                 26%

Birmingham       12%

Manchester       32%

Leeds                 18%

Nottingham       3%

Aberdeen           107%

That's a good old spread of fortunes for house prices in one country, over the last decade, isn't it?  All over the place.  And does not fit with the idea that prices in UK double every 10 years at all.

What's the point then to looking at that?  Why not just await the next house price prediction from the experts?

Outside London, Aberdeen showed an outstanding growth of >100% from 2004-2014.  However, a report from 10 years ago in 2004 would have shown Aberdeen to have been [one of] the worst performers in UK from 1994-2004. The city had mainly flatlined during that period, as the graph below shows.   

So what would have been one of the worst performers in 2004, becomes the star of 2014. Who knew?  [We also see in the above graph that Edinburgh was one of the stars of the previous 10 years from 1994-2004, then one of the worst in the next 10 years].

Next of interest is that  many big centres show a really low 10 year growth up to 2014 [Edinburgh - 28%, Leeds 18%, Birmingham 12% Nottingham 3%].    Is it within these poor-performing areas where the value actually lies in the next 10 years?  In other words can we say:

IF SOMEWHERE HAS GONE UP IN LAST 10 YEARS, SELL.  IF IT HAS FLATLINED FOR 10 YEARS THEN BUY

Well, maybe we can.  Some cities will fair better or worse than others regardless of their past performance of course.  Big things can happen like oil price goes up [Aberdeen's rise must be down to this to such a large degree] or jobs can go at a major employer in an area which really shakes the market down.  But all things being equal, if we believe in a cycle to UK property then looking back 10 years is a good guide anyway to the next 10.  Seems nice and simple...a monkey could have come up with that.  Perhaps even a drunk one.

SO WHAT?

I just think this kind of proves that "past performance is an indicator of future growth".  In other words, over the 10 year view, places which post low or very low growth could well be the places to look to for the next 10 years.  And vice-versa. Just a re-work really of the 20 year property cycle ideas, but down to city level with the Nationwide data.  

What Could You do With This Idea?

A new-comer to UK, not knowing any of the cities, would look to this data in this way, pick out say 10 big cities which have had lowest growth over last 10 years and then do the due diligence on the ground in terms of micro locations and rental demands and levels / trends and then buy.  If an area which has gone down for a decade or flat-lined is actually giving good yields which are highly sustainable through demand, then seems like a good idea to buy as many of that as you can get your hands on.  You will find cities that have gone down for 10 years or flat-lined, and where it is still difficult to get a good yield or a reliable yield through good tenant demand.  Leave that stuff alone.

So using this data could really help focus on where to buy, when to buy it and how much it should increase in the usual 10 year upturn for that actual city and so also a key flag for the time a sell.

And London????

Ok, London.  It is an odd one here, the global recession has had such an odd effect on this city.  Well, the earnings ratios of 7.7 or so  would lead you to think that a pull back to a ratio of 6.0 or so is due in the city at some point soon, a 25% price correction, based on the other 2 pull backs which occurred in London in the cycle since 1989.  Seems odd that London is going up so much at the moment, by every measure.  You would think that tightening of lending criteria, interest rates creeping up and £ getting stronger will act now in London.  I have been saying that since 2003, to be honest, so what do I know.

HERE ARE MY GUESSES - FOR WHAT IT IS WORTH

And so, the monkey is sufficiently imbibed and has some darts in his hand. The 5 and 10 year prediction ranges are then, based on the ideas of past pricing and also of current price to incomes in each area:

City                                    2019 Prediction            2024  Prediction 

London                                -25 to -5 %                             25-45%

Edinburgh                            25 - 35%                                 60-80%

Cardiff                                  25 - 30%                                65-80%

Birmingham                         30-40%                                   70-90%

Manchester                         30-40%                                   75-100%

Leeds                                  35-45%                                  80-120%

Nottingham                         20-35%                                   70-110%

Aberdeen                            5-15%                                     15-35%

A bit of a cop-out to use ranges perhaps, and some seemingly big numbers in there.  Is the monkey "bullish" with his darts?  I think not. Looking at Nottingham for example, and putting the last 10 years growth of 3% in there, this is just guessing at a growth of around 85% over the total 20 year period from 2004 to 2024.  UK prices actually very rarely grow this slowly over a 20 year period, if ever.  But the hangover of ultra-low interest rates, and low wage inflation probably means we are into a lower growth era than before.

What Next?

Keep this blog link.  Look up the prices and compare in 2019 and 2024 the above "drunk monkey" throws and the expert predictions.  And then, if the ranges above were closest, perhaps buy a monkey a drink and ask them where to buy some property.