Over the past few months I have been exploring what actually drives UK House Prices. In developing some mechanical non-emotional datasets I’ve come to the conclusion that the driver is actually UK House Affordability. That said while Affordability drives the housing market I personally only want to buy a house when it is at a sensible Valuation. Therefore from here on in I intend to monthly monitor two key UK House metrics:
Before we look at the metrics let’s first look at the key pieces of data I am using to assess both Value and Affordability:
Unfortunately, the Average Weekly Earnings dataset limits this analysis to January 2000. I however want to look at longer term trends to try and judge where fair value may be and even what P/E lows we could expect going forward. To get an indicator of this I use an older similar dataset which was discontinued by the ONS in September 2010. This was the Seasonally Adjusted Average Earnings Index (AEI) for the Main Industrial Sectors. This dataset goes back to 1990 which is sufficient to take us back through the last UK property bust. I then convert the Average Weekly Earnings dataset to an index and overlay both on the chart below. This shows that today we are nowhere near fair value and so I stay in rented accommodation.
Before you start flaming me in the Comments below I know that there are other factors such as reports that Banks have tightened up there lending criteria including not offering very high lending multiples. This may be the case however the Financial Times reported on the weekend that while we no longer have 125% mortgages we do still have high loan to value multiple deals available out there. Have no deposit and there are still 5 mortgage offers available which astounded me given the world we live in today. Have only a 5% deposit and you still have 67 deals to apply for.
To me these charts in conjunction with government house price backstops in the form of the Mortgage Rescue Scheme and Support for Mortgage Interest to name but two all implies that we won’t see any substantial house price falls in the near future. These will only come if interest rates continue to rise and/or the government runs out of other people’s money.
As always DYOR.
Note: Apologies for the poor quality of today’s charts. I have temporarily lost access to my normal laptop so had to improvise to bring you today’s post.
- I will monitor UK House Affordability which will hopefully give me some insight into whether house prices will be increasing or decreasing in the foreseeable future.
- As I remain in rented accommodation and intend to buy when prices are fairly valued I will also monitor UK House Value. This will hopefully give me a sensible buy point to ensure I don’t lose money on the purchase.
Before we look at the metrics let’s first look at the key pieces of data I am using to assess both Value and Affordability:
- UK Nominal House Prices. I have consistently been using the Nationwide Historical House Price dataset for a lot of previous analysis and so will stick with it. August 2012 house prices were reported as £164,729. Month on month that is an increase of £339 (0.21%). Year on year sees a decrease of £1,185 (-0.72%).
- UK Real House Prices. If we account for the devaluation of the £ through inflation (the Retail Prices Index) we see a very different story. Month on month that nominal increase turns into a decrease of £263 (-0.16%) and year on year that decrease grows to a larger £6,031 (-3.66%). In real terms prices are now back to those seen in March 2003.
- UK Nominal Earnings. I choose to use the Office for National Statistics (ONS) Average Weekly Earnings KAB9 dataset which is the seasonally adjusted average weekly earnings of both the public and private sector including bonuses. July 2012 see earnings at £471. Month on month that is an increase of exactly £0. Year on year the picture is not much better with an increase of £6 (1.27%). With inflation (the Retail Prices Index) running at 3.2% over the same yearly period purchasing power of those that work continues to be eroded.
- UK Mortgage Rates. The proxy I use to monitor mortgage interest rates is the Bank of England dataset IUMTLMV which is the monthly interest rate of UK resident banks and building societies sterling Standard Variable Rate (SVR) mortgage to households (not seasonally adjusted). August 2012 saw this reach 4.26% which month on month is an increase of 0.02% and year on year is an increase of 0.16%. So while the Bank of England holds the Bank Rate at 0.5% out in the real world we are seeing mortgages start to cost more, even if it is happening very slowly.
UK House Value
The stock market uses the Price to Earnings Ratio (P/E) as a possible valuation metric. I choose to use the same metric to assess housing value and show this in my first chart below. For Price I use Nominal House Prices and for Earnings I use the UK Nominal Earnings multiplied by 52 to convert to Annual Earnings. This shows that today we are sitting on a P/E of 6.7 which is a small decrease on last months 6.8. While being a long way off the peak value 8.3 we are also still a long way off of the 4.6 seen in January 2000.
Click to enlarge
Unfortunately, the Average Weekly Earnings dataset limits this analysis to January 2000. I however want to look at longer term trends to try and judge where fair value may be and even what P/E lows we could expect going forward. To get an indicator of this I use an older similar dataset which was discontinued by the ONS in September 2010. This was the Seasonally Adjusted Average Earnings Index (AEI) for the Main Industrial Sectors. This dataset goes back to 1990 which is sufficient to take us back through the last UK property bust. I then convert the Average Weekly Earnings dataset to an index and overlay both on the chart below. This shows that today we are nowhere near fair value and so I stay in rented accommodation.
Click to enlarge
UK House Affordability
I believe that the Average Joe out there doesn’t have any concept of house price value and instead is just interested in how much he can borrow from the bank. I track Affordability using a dataset I have created which I call the UK House Affordability Ratio. I define this as the Ratio of Average UK Monthly House Repayments to Average UK Earnings at the point of the mortgage being granted.
Let’s first calculate the Average UK Monthly House Repayment. This is calculated by taking the Nationwide dataset, the Bank of England’s SVR dataset along with assuming a 20 year, 90% repayment mortgage (the actual value isn’t overly important as it is held as a constant through the dataset for comparison purposes) and is shown in the chart below.
Let’s first calculate the Average UK Monthly House Repayment. This is calculated by taking the Nationwide dataset, the Bank of England’s SVR dataset along with assuming a 20 year, 90% repayment mortgage (the actual value isn’t overly important as it is held as a constant through the dataset for comparison purposes) and is shown in the chart below.
Click to enlarge
We can then ratio this with Average UK Earnings to arrive at the UK House Affordability Ratio which is shown below. Remove the credit boom and affordability seems almost range bound between 0.40 and 0.45 (represented by the orange lines).
Click to enlarge
Before you start flaming me in the Comments below I know that there are other factors such as reports that Banks have tightened up there lending criteria including not offering very high lending multiples. This may be the case however the Financial Times reported on the weekend that while we no longer have 125% mortgages we do still have high loan to value multiple deals available out there. Have no deposit and there are still 5 mortgage offers available which astounded me given the world we live in today. Have only a 5% deposit and you still have 67 deals to apply for.
To me these charts in conjunction with government house price backstops in the form of the Mortgage Rescue Scheme and Support for Mortgage Interest to name but two all implies that we won’t see any substantial house price falls in the near future. These will only come if interest rates continue to rise and/or the government runs out of other people’s money.
As always DYOR.
Note: Apologies for the poor quality of today’s charts. I have temporarily lost access to my normal laptop so had to improvise to bring you today’s post.
This is all good stuff but I will observe that, for people looking to buy, it might be more useful to check geographical focus, particularly in London. Not sure how feasible this is, but regional cyclicity may have its part to play.
ReplyDeleteHi SG
DeleteI agree that an analysis and then comparison of an inside M25 vs outside M25 (or similar) could reveal some intersting trends. Unfortunately, I don't think that the freely available datasets that I'd need are out there.
Thinking aloud:
- I'd need a house price dataset. I could probably switch to the Land Registry data which would certainly give me "London". The question is how would I calculate the outside M25 data? I can get the price for each of the other regions but to create an average for all regions I'd also need the volume per region which doesn't seem to be published.
-I'd need earnings data for both regions which following a quick search doesn't seem to be available. This I think would possibly show some very revealing results were it to be available.
- Mortgage Rates. It would probably be a reasonable assumption to use the same SVR for both inside and outside.
So if any readers can help me with a dataset for prices and earnings in and out of London I'd be willing to crunch the numbers.
Cheers
RIT
I'm not sure you will ever buy a house
ReplyDeleteThey are based on affordability you are right, but they always look expensive compared to renting
The reason is that most people value security of tenure and being able to have a home the way they want it over the flexibility of being able to move at short notice
This doesn't really fit into your graphs and spreadsheets I think
I say this with a housing history going back to the 80s
Another flaw in your analysis is as the previous poster has written, there are really two markets in the UK right now the London market and the rest
I think around 25% of the "value" of all the housing in the UK is in greater london and another 25% is in the surrounding counties
Hi Anonymous
DeleteWhen running my analysis I am not considering rents. I am looking at the ability for somebody to service a mortgage which is what I believe drives prices.
Cheers
RIT
Anonymous -
ReplyDeleteNope, there are not two markets, and SG did not claim that. Stop twisting people's words.
It's all the same market, and prices in London, whilst naturally higher, are falling slowly just as they are elsewhere.
You've let yourself be fooled by the averages, which are of course skewed by the occasional oligarch buying a mansion in Belgravia from time to time.
Remove those guys from the stats and you'll see: not a good time to own property in London or elsewhere.
Hi Steven
DeleteThanks for jumping in with some thoughts. Given you're a Chartered Surveyor you are clearly living this everday and are much closer to it than I am. I and I'm sure Retirement Investing Today readers would love to hear more about wht you are seeing out there? Particularly given your last teaser paragraph.
Cheers
RIT
Good stuff... I'd love to see London broken out, too, though that's because I follow that mad market most closely! ;)
ReplyDeleteHi TI
DeleteAgree with you. It certainly is a mad market.
Steven you said: "You've let yourself be fooled by the averages, which are of course skewed by the occasional oligarch buying a mansion in Belgravia from time to time."
ReplyDeleteThat sounds like two markets to me?
I happen to me in the market for (non-oligarch) property in London and trust me, these properties (to my surprise) are actually selling for 20-40% above "peak" 2007 prices - you might not like but it is true
(Yes, I am tracking some of these through to the land registry sold prices)
Anonymous
Hi RIT,
ReplyDeleteA great post, and the more I read of your site, the more impressed I become with your research.
I am interested in your perspective of "value" however....
You say;
"I personally only want to buy a house when it is at a sensible Valuation".
Surely the only real way to maximise "value" is to allocate the smallest percentage possible of your lifetime earnings in order to pay for your lifetime housing costs.
With such a goal in mind, the key variables over a lifetime would be rental costs, mortgage interest costs and of course the 'sticker price' of a house.
I can't help but feel you perhaps overestimate the importance of one of those components, the house price, in reducing lifetime housing costs.
Given that the fall in nominal house prices has been relatively small on average, and that nominal prices look unlikely to fall further in any meaningful way, it would seem the real savings are to be made exploiting the once-in-a-lifetime low interest rates on offer with the mortgage deals available to almost anyone pre-2008, or sadly to a somewhat lesser extent today.
Particularly as rents seem to be soaring to new record highs on a monthly basis at the moment, dramatically increasing the lifetime housing costs of Generation Rent.
Fixating on finding 'value' in just one of the three components to lifetime housing cost reduction, while overlooking or underestimating the importance of the other two, would seem to me to be a mistake.
So I'd be interested to read your thoughts on this matter.
Cheers
A1
Hi A1
DeleteI very much agree with your statement that the game is "... to maximise "value" is to allocate the smallest percentage possible of your lifetime earnings in order to pay for your lifetime housing costs."
I feel that I am actually playing that game through a few techniques:
- A key one that you may have missed is that I am taking advantage of the one small benefit of renting. That is that I can move quickly (a positive and negative of AST's) for very little cost. Therefore today I am renting a place which is much smaller than I intend to buy. I am also living in a development which is fine today but I believe will not be a very pleasent place in the future. My plan is to either buy when/if sensible prices prevail or move to another rental when the degradation occurs to a level I am not happy with. This means that my rent today (assuming no capital tied up in property as I have that deployed elsewhere which I'll cover in a second) is less than what I would be paying in interest on a purchased property (I wouldn't have been able to buy pre-2008 so would have missed those lifetime low rate trackers). As written in a previous post this strategy also then gives secondary benefits in the form of reduced heating, lighting and council tax which also go into the Portfolio.
- The capital not deployed to housing is providing a return elsewhere. Year to date (to August) the Nationwide suggests that nominally the "typical" (Note: I know we still need to work through these indices in detail and I very much look forward to your further thoughts whether as a post or just Comments over the coming days and months) is up 0.6%. To enable a comparison let's say I deploy 25% of my assets to a house deposit. I would actually deploy more than this but this will give a best case ROI for a sensible mortgage rate. A 25% deposit therefore provides a 0.6% x 4 = 2.4% return on capital invested. In comparison to end of August (so roughly same period) my Retirement Investing Today Low Charge Portfolio is up 7.7%.
So given my rent payments are less than what my mortgage payments would be (I get the additional benefit of not having to pay for repairs which I haven't factored in) and I'm getting a better return on capital then I feel I am possibly slowly winning the war. If I wasn't prepared to live in a smaller property with the intention to move as required then of course it could be a very different conversation we are having.
You might also say I have selection bias. Therefore let's have a quick look at the last few years when I would have had sufficient funds to deploy to housing:
- 2011. House prices up 1% so a 25% deposit would have given a 4% ROI. My portfolio return 2.6% so I lost that one by 1.4%.
- 2010. House prices up 0.1% so a 25% deposit would have given a 0.4% ROI. My portfolio return 13.6% so I won that one by 13.2%.
- 2009. House prices up 5.9% so a 25% deposit would have given a 23.6% ROI. My portfolio return 22.8% so I lost that one by 0.8%.
Therefore looking 2009 to present am I "crudely" ahead or behind. Let's assume my capital invested was £100:
- Current strategy £100 x 1.228 x 1.136 x 1.026 x 1.077 = £154
- Deploy 25% to housing £100 x 1.236 x 1.004 x 1.04 x 1.024 = £132
So even assuming I bought back at the start of 2009 (prior to the commencement of this blog) I feel capital wise I am still around 16% ahead by being out of the market. In that time I also haven't yet had to pay the Government Stamp Duty which is compounding, have saved the delta between interest paid and rent, have had to pay no repairs and have saved utility costs. All of which is now deployed in my Portfolio and compounding nicely.
Unless I am missing something I therefore feel that I am at least on the right path to achieving what you highlighted was the goal.
Thoughts?
Cheers
RIT
I think you've raised a number of interesting points.
ReplyDeleteYes, I would agree if you can reduce your rent expenses then a comparison against buying can certainly stack up more positively.
The young person living with their parents for free, to use the most extreme example, would have a clear financial advantage over someone paying market rent.
Likewise, a young person living in a house with several friends could achieve significant savings versus the cost of living on their own.
But where I'm not entirely convinced your argument makes quite so much sense, is when a person is choosing to rent a smaller place versus buying a larger one.
Surely in such a circumstance, the real comparison would be in terms of renting a smaller place versus buying an identical smaller place. Particularly when, as your posts suggest, you may be/have been in that level of housing for somewhere between 5 and 10 years. In which case, the comparison is much simpler, and involves rent versus mortgage interest versus capital gain/loss, on an identical like for like house, rather than comparing different segments of housing.
As an example, the current average rental yield is circa 5.5%, the current average mortgage rate is circa 3.4%, so the gain from buying versus renting is circa 2.1% per year. (When comparing like for like properties.) So unless prices are falling by more than 2.1% per year, renting (by my crude way of thinking anyway) makes little sense, even if prices are falling nominally, so long as the cumulative nominal falls do not exceed the cumulative difference between rent and mortgage interest over the time you would spend renting.
Of course, renting in the short term can certainly make sense for a variety of reasons, the flexibility to move across the country for a new job, to try out a new town/area, etc.
But when you start talking about a 5-10 year timespan, I very much have my doubts about whether renting can deliver any kind of financial advantage in reducing lifetime housing cost versus buying. Even allowing for the nominal terms price falls we've seen to date, on average.
Cheers
A1
Hi A1
DeleteI guess everyones situation is different and your example could be appropriate for others. I am happy to rent my current place for a few years however I would never buy here. If I bought this place then I'm up for stamp duty, maintenance costs and then real estate/solicitor fees when I sell. I would also have to deploy capital (deposit) plus be geared on an asset which I believe will be less pleasent (and hence not hold value) in the future. Your example doesn't work for me but to keep the debate going though let's run your example for my situation.
Using Zoopla as a proxy for value my Landlord is currently achieving a yield of 4.2%. Cheapest fee free (to make this analysis easy) mortgage seems to be a 2 year fixed with an AER of 3.3%. So comparing like with like the arbitrage is closer to 1.1%. So providing house prices, after netting off all the costs mentioned above, go up by about a third (assuming a 30% deposit which is required for this deal) of the rate of my alternate investments I guess I could get slightly ahead.
So it's really just a geared play on real estate versus a balanced portfolio of other assets (which includes commercial property). Only time will tell which is the better choice. However given the FTSE 100 and ASX 200 (which together are about 40% of my portfolio) is according to my analysis about fairly valued and real estate is according to the above over valued I know where I would (and I guess do) place my chips.
Maybe my situation is just unique but I still believe I am on the road to allocating "the smallest percentage possible of your lifetime earnings in order to pay for your lifetime housing costs." I guess only time will tell.
Thanks for the great debate.
Cheers
RIT
@A1
ReplyDeleteI agree with the general thrust of your argument but your numeric analysis is just plan wrong :(
You say: "As an example, the current average rental yield is circa 5.5%, the current average mortgage rate is circa 3.4%, so the gain from buying versus renting is circa 2.1% per year."
My complaint is two fold:
- you ignore key other variables, namely: annual house value change for owned property; repair costs on owned property; and annual rental cost increases
- you assume that the current mortgage rate is never going to change over what is typically a 25 year term; in reality central bank interest rates are at a 300 year low (I think the 20 year average for mortgage rates is something like 7-8%
Anonymous
I think "wrong" is a bit of a stretch.... ;)
ReplyDeleteAlthough I'll certainly concede that it is very much a simplification for the purposes of a general discussion.
So to address those points....
- you ignore key other variables, namely: annual house value change for owned property; repair costs on owned property; and annual rental cost increases
Well, yes. In the interests of keeping the conversation simple enough for general points to be discussed, I have indeed restricted the variables to mortgage interest, rental yield and capital gain/loss.
"Annual house value change" is only really relevant WRT the actual transaction price at the point of buying and selling, but I addressed that point within the discussion of capital gain/loss.
In addition to that, as you rightly point out, you have maintenance, transaction costs, letting agent fees, moving costs, etc.
But in my opinion, over a 10 year timescale the initial transaction costs and ongoing maintenance for a small cheap place would not be markedly different than the letting agency fees and moving costs for a typical renter that moves every 6-18 months.
- you assume that the current mortgage rate is never going to change over what is typically a 25 year term; in reality central bank interest rates are at a 300 year low (I think the 20 year average for mortgage rates is something like 7-8%
Oh not at all.
I'm certainly not assuming that the rate will stay this low for 25 years. But I am noting that the rate has been exceptionally low for almost 4 years, and looks likely to remain this low for quite some time yet.
I would therefore suggest it's not unreasonable to note that taking advantage of perhaps 8 or 10 years of record low interest rates can have as big an impact on reducing lifetime housing costs as achieving a significant reduction in the house price.
In an ideal world of course, we'd be able to do both.
But as it seems unlikely that prices will fall significantly from here unless rates rise both markedly and prematurely, (as raising rates after the recovery is embedded will likely not achieve such results... after all a growing economy, reducing unemployment, and increasing wages are not normally associated with falling house prices)then I fear that those seeking 'value' only in terms of the price of a house will actually find that they end up paying the same as they would today. But miss out on the opportunity for lowering lifetime cost because they miss the once-in-a-lifetime low rates.
And combined with paying that rent for a decade or more, will have increased their lifetime housing costs markedly.
That's what is so pernicious about the current situation.
Prices have fallen since 2007, but it's done precious little good to those who now cannot get mortgages and avail themselves of the currently low rates.
I'd suggest that for many/most of those people they'll eventually be able to buy at a lower 'sticker price' than they would have pre-2008, but only at the expense of markedly increasing their lifetime housing costs through being forced to rent for so long, and then buying into rising rates when the recovery takes hold.
We're storing up problems for the future here, as an entire generation are financially disadvantaged, forced to buy a large part of a house for their landlord instead of devoting that expenditure to buying one for themselves.
It's not a good place for society to be, in my opinion anyway.
Cheers
A1
Great post, thanks.
ReplyDeleteWhen I use P/E to value shares, I look at price of the share & company earnings per share. To get a comparable P/E for house prices, would it be better to use the price of a house (as you have) and the earnings for the house, such as its ability to earn rent (or to save you rent), rather than average wages?
A useful second line on the house value chart might be a comparison to the next most likely asset class; stock market returns. This could be a useful yardstick for the "opportunity cost" discussion in the previous comments.
Cheers, Ric
Hi Ric
DeleteYour proposal could indeed work but it's not the method I have chosen to use in this case. Have you run this analysis? I'd be interested in the results.
To expand on my method a little to hopfully further clarify why I do what I do. The reason I have used the method I have is that if you think about the affordability vs value metrics above the big difference is that the value metric does not account for interest rates. Otherwise they are in fact pretty similar. Therefore in a crude way my value argument is almost an affordability index with average interest rates (almost smoothing the typical business cycle like I do for the stock market with my PE10). Therefore as I believe it is affordability which drives the market I'm ok with the P/E used.
Thanks for the suggestion of adding stock returns to the chart. If your interested I also run that analysis (both PE and PE10) for the FTSE, ASX and S&P. A link to the datasets can be found in the top left sidebar.
Cheers
RIT
@ A1
ReplyDeleteOkay so now I'm being pedantic. You say:
"I would therefore suggest it's not unreasonable to note that taking advantage of perhaps 8 or 10 years of record low interest rates can have as big an impact on reducing lifetime housing costs as achieving a significant reduction in the house price."
According to my FT the best 10 year mortgage rate on the market is 4.79% (you need a 25% deposit of course). In your first post you say that the typical yield on a BTL is 5.5%
There isn't really that much of an arbitrage there
Anonymous