Very cool, if true

In terms of making the house market viable again though, the ratio of salary to house price is a reasonable value if (and who knows what they're really thinking at the mo) the banks have it in their heads that they won't lend at more than 3.5x or 4x a salary to avoid bad debts.

And if banks won't lend to risky customers, no matter what the interest rate, then the ratio will have to drop for houses to become affordable.

From:

drplokta

Actually, the banks lending at 3.5x to 4x salary indicates that they are taking lower interest rates into account. Back in the late 80s, when I bought my first house, it was 2.5x to 3x.

The main problem at present is that you need a deposit of 20% or 25% to get a decent mortgage.

From:

The deposit is indeed my worry at the moment, because even if I can rent my house in London out for a sum that will cover the mortgage, I don't have a lump sum deposit to put down on a place here in Cambridge.

And it will probably take me more than six months for the difference between what I paid in London and what I'm paying in rent to generate enough money to pay for the house repairs in London ... so I can't start saving for a deposit until the second half of this year ... but if I sell off a lot of stuff (I don't need it!) and cut back on expenses (like conventions!), maybe I'll be able to look at buying something in second half 2010. Or a lottery win would be nice!

From:

the-locster.livejournal.com

Yeh I guess it would be worth a look. I actually started out looking at property price and interest rate correlations back in 2001ish. Off the top of my head I'd say my main gripe with such a metric is that interest rates do tend to change very abruptly, so that what is affordable today might not be tomorrow due to government and central bank rate setting; which in turn is an artificial manipulation of markets. In fact in the global sense it is possible to define a 'true' interest rate, albeit one that we can't ever know for sure, but then again we don't know what the official rate will be for sure tomorrow and thus effectively today. Yes you can take out a fixed rate loan, but that just hits the bank or yourself when the rate changes, thus if everyone took out such loans then the rate becomes unsustainable either for the banks or the debtors - the *true* rate always wins in the global(mathematically speaking) case, as we are witnessing right now.

The crux of my argument is that official rates are not indicative of the *global* or *true* rate, which, getting back to basics, is the rate at which capital is able to produce new capital and goods(+money supply rate). As such you might take the view that the official rate is effectively a random variable that at best only loosely correlates with the true rate.

Another way of looking at it is as short term chaotic fluctuations in official interest rates versus what is clearly a slower cyclical dynamic observable in the graph. That is, the main characteristics of the slow dynamic are perhaps largely independent of the short term dynamics of interest rates by way of a multitude of feedback mechanisms that we barely understand + the reactionary dynamics of politicians that are entirely predictable in these situations.

From:

drplokta

Also, the house price data set used in the graph is clearly either the Halifax or Nationwide one, as it's showing prices peaking under £200,000, which is wrong. First, Halifax and Nationwide track the price of a "typical house", and their figure is more like a median than a mean, so you shouldn't compare the data against mean earnings. Second, they adjust out improvements in the housing stock, so they assume that the housing stock does not change in composition or quality over time. Go and get some decent figures for the actual mean house price from the DCLG, or the Land Registry if you can dig deeper than their new index (which uses a geometric mean rather than arithmetic).

From:

There's not much point telling me to improve it, I'm merely copy/pasting from elsewhere, and I certainly don't have the time to dig up figures myself. If you have access to a better graph I'd love to see it though.

From:

It's the Nationwide data (it says so on the page linked to for the chart)
Their methodology is here

It's basically an adjusted and "cleansed" set of values from the mortgages in the approval stage. It excludes remortgages, right to buy at discounted values and houses that are outliers (for example too large or too small for the average house in that area). And they then "balance" the data for North/South, seasonal, mix of flats to houses being mortgaged etc.

So it's an interesting figure ... but at best it will indicate a trend and at worst it's measuring the wrong thing and then applying fiddle factors that make it worse than useless by being actually misleading.

From:

lebeautemps.livejournal.com

I'm not thick, but could you explain what a "Multiplier" is on this graph?

From:

cdave

Ratio of house price to earnings

From:

average earnings x multiplier = average property price.

Not quite the same as "how many times your salary a bank will lend you a mortgage for" but similar (the former doesn't account for deposits).

From:

Wouldn't net earnings be a more useful figure?

From:

Possibly, but as Andy said above, he's just cutting and pasting the graph from elsewhere. You could also argue that net earnings vs average value of mortgage loaned rather than average property price might indicate different things too.

Given enough stats we can demonstrate pretty much any result we want to :)

From:

True.

But gross earnings don't seem to actually mean very much. Net earnings indicate the amount of money people actually have, gross earnings indicate the amount of money people might be given except a big chunk is taken away first. AFAICT gross earnings are always a fairly pointless statistic - ideally even job adverts should not advertise gross salary.

If I get paid X but only ever see Y (which is less than X) why even tell me about Y?

This becomes even more nonsensical when someone is trying to show how affordable something is based on X (a number which has nothing to do with how much money I have to spend).

Edited Date: 2009-01-26 09:35 am (UTC)

From:

Oh I agree, unfortunately average net earnings aren't as easy a figure to get hold of I suspect - certainly not one you can calculate based simply on the tax bands and the average gross income figures as there's capital gains tax, benefit income, family tax creidts etc etc etc to consider.

And whilst not as meaningful on a personal level in such things as job adverts, average gross incomes are probably still a half-decent indicator of general overall national wealth in this situation.

From:

Gross/Net conversion isn't simple though, and can vary significantly between people depending on benefits, perks, etc.

One person I know is technically on minimum wage, as 95% of his salary goes straight into his pension, I plough some of mine into an untaxed sharesave scheme, soem people I know get childcare vouchers, etc.

From:

Still, most people don't do that.

From:

What, have childcare vouchers or sharesave schemes?

Pretty much everyone at the company I work for is getting _some_ kind of benefit that effects tax.

From:

I don't think these things effect income as much as general taxation, that's my point. Yes we can't get a perfect figure, but if we subtract the main bit, which is general taxation - we get a much closer figure that corresponds to how much someone could actually spend on a house.

No system is perfect, but when some people are having a big chunk of their income taxed at 40% while others are getting 10% or whatever it makes a big difference.

From:

If you're getting your income taxed at 10%, you're not likely to be buying a house.

For the rest, assume between 30 and 40%. If you have an average income of around £22k, then the "Y" figure is about £14k.

Over "recent" historial times, that approximate percentage has been fairly steady (a few percent either way) so the income to house price multiplier will be pretty much identical, just the entire lne shifted up or down a little, in my estimation anyway.

Of course you could look at the fact that "Median earnings of full-time male employees was £521 per week in April 2008; for women the median was £412."
http://www.statistics.gov.uk/cci/nugget.asp?id=285

and come up with an interesting graph of male & female income over the years vs house prices ... not sure what it would "prove" but it would be interesting!

And of course none of those income graphs show the direct effect of housing ladder uplifts ("I bought at 100k and 3 years later I sold at £180k") ... which pushes house prices up mostly through requiring larger mortgages (or so I'd claim) and thus requiring higher multiples and riskier mortgages, until the whole thing collapses.

When I bought my house eleven years ago, I bought at three times salary. Ten years later it was "worth" 2.5x what I'd paid, but my salary had only risen about 30%, so that I would not have been able to afford my house at FIVE times my increased salary. Even with the recent drops, it's still looking like I should be able to expect a buyer (if there was one) to pay twice what I paid for the house, which is still more than three times my salary (still more than four times my salary in fact!)

From:

Even with the recent drops, it's still looking like I should be able to expect a buyer (if there was one) to pay twice what I paid for the house, which is still more than three times my salary (still more than four times my salary in fact!)
And that's what I'm questioning - whether people really can expect that, or if it's an illusion and we're all going to have to wait for prices to drop significantly.

drplokta is right about taking interest rates into account though - the question is whether we can expect them to stay incredibly low for a while...

From:

Not sure about interest rates, not sure about inflation either.

And of course if house prices do start to drop down to where my house comes back to three times my current salary, I'll be in about 60k of negative equity (which is *way* more than a year's salary before or after tax!) and a large number of other people will find themselves in a similar state (I'm guessing potentially millions of people, though it might only be tens or hundreds of thousands) thereby exposing banks (and now the British taxpayer) to even more risk.

Sure it makes it easier for people to buy, but it makes it harder on those of us who have already bought. And yes, it's my own fault for getting into debt and spending money I thought I "had", I'll admit that. I'm not asking other people to pay for that (well, if I could I would, but I can't!)

From:

Oh, I'm not saying that massive price falls are a _good_ thing - I'm just wondering if they're likely. Taking into account the lower interest rates, it seems unlikely to me that they'll fall to 3x. 5x seems perfectly possible though.

From:

Phew, at 5x I could afford to buy my hosue :-)

And of course the "benefit" of a tighter mortgage market at the moment is that it means it's harder to get a mortgage as you need a sizeable deposit, so there are more people needing to rent, so I should be able to find someone to rent my house :-) (fingers crossed!)

From:

I hope that "phew" wasn't you thinking I know what I'm talking about...

From:

the-locster.livejournal.com

I just want to add that I entirely agree with this point. In fact an even better metric would be to subtract tax along with some every day costs of living, such as the bare minimum required to buy food and clothes. What you have remaining you can then choose to spend on higher quality food/clothing (but you don't have to) and accomodation - in the form of rent or a mortgage.

However as others have pointed out this data is less easy to get hold of and unravel, although we could get closer to the truth by simply estimating an average tax+minimal living cost and subtracting that. E.g. if we just took away 20% of gross earnings we'd be closer to the truth.

The resulting graph would (as I think andrew said) be more or less the same shape but with the multiplier(ratio) line shifted up, and perhaps slightly flattened because as earnings fall towards those bare minimum living costs your 'free' income approaches zero, giving the multiplier a boost.

This also means the salary multiplier used when applying for mortgages is a poor metric, and in fact what you often see is the banks quoting different multipliers for different salaray ranges - which is just a hack/correction required because they didn't subtract tax and living costs before applying the multiplier.

I wouldn't say it's entirely nonsensical because it does show how prices vary cyclically as a proportion of how much money people have to spend on accomodation. You might expect such a multiple to be fairly consistent, but it isn't.

From:

And

drplokta has pointed out previously, that in price declines such as the one we're in at the moment, people that don't have to sell, don't sell, they hold on. So people with money (and therefore probably expensive houses) keep the house rather than sell it cheaply. People at the lower end who *have* to sell (losing job and having to move elsewhere or sell to clear debts) plus repossessions mean that the average selling price drops a lot, but the number of properties being sold also drops (and with the current mortgage market, there are far fewer buyers able to get a mortgage at the moment, due to deposit requirements etc.)

This makes figuring out a valid "average house price" really difficult. All you can really give is the current "average selling price" rather than for the housing stock as a whole. Again as

drplokta so cogently puts it, housing stock is improving/ A lot of new build and a fair chunk of older cheaper property has been demolished/replaced or renovated during the property boom.

And again for "selling price", some of the flats around west London are being sold on a "we'll pay 25% of your flat price" so for a £200k flat, you pay £150k and the builder "pays" themselves the other £50k. This means the land registry sees a sale at £200k but the actual selling price is £150k ... or the deals to buy 1/6th or 1/8th of the flat (so paying £25k on a "£200k" flat) and paying rent for the rest. Or even the deferred "buy half" scheme where you pay for 50% and the deal is when you sell, you give 50% of the sale price to the builders, they assume prices will rise before you sell and so it's an investment for them, gets an empty property off their books, gets some cash flow and keeps the apparent price propped up.

Edited Date: 2009-01-26 09:57 am (UTC)

From:

lebeautemps.livejournal.com

Thanks.

From:

lebeautemps.livejournal.com

And I wish I understood more about economics - where has the multiplier data come from? Surely if its a question of av cost/av salary, then the multiplier should be largely following the curve...?

From:

The multiplier is at linear scale, the other curves have logarithmic scale.

From:

drplokta

Here's another graph from the Nationwide (so it's still using their slightly duff house price series), showing initial mortgage repayments as a percentage of gross income (the red dotted line). It shows a much less extreme peak, much smaller than the one in the late 80s.

From: