Who'd vote for lower house prices? Not many...

Graham_Devon

ukcarper wrote: »

I think I've already posted that if you take into account difference between mean and median the earnings to house price ratio is 4.77x

I think that's a hesitant yes

ukcarper

Graham_Devon wrote: »

I think that's a hesitant yes

Take it how you like but it looks like if prices fall 16% they will be below long term average in repect to median earnings. Not the 50% or more claimed by many on here.

wotsthat

Graham_Devon wrote: »

Ultimately, I have just categorically proven that house prices are about 40k overvalued. But if you ignore the mean wage moving away from the median wage and pretend most people are on 37k a year....your bubble remains intact.

Can I just check something? Do you understand that if you plot average full time male earnings OR all wages against house prices VS time both will be basically the same? I think you've spent a long time proving you don't. Honestly Graham - you haven't categorically proven that house prices are £40k overvalued.

The reason the graph is relevant is because good historical data is available and maintained from a reliable entity. It's not a conspiracy to try and make the earnings ratio look lower.

What graph would you trust - ONS all earnings vs. Land Reg? Or are we meant to 'just look out of the window'?

ukcarper

wotsthat wrote: »

Can I just check something? Do you understand that if you plot average full time male earnings OR all wages against house prices VS time both will be basically the same? I think you've spent a long time proving you don't. Honestly Graham - you haven't categorically proven that house prices are £40k overvalued.

The reason the graph is relevant is because good historical data is available and maintained from a reliable entity. It's not a conspiracy to try and make the earnings ratio look lower.

What graph would you trust - ONS all earnings vs. Land Reg? Or are we meant to 'just look out of the window'?

I do agree with Graham to some extent that the relationship between mean and median can vary and median would be a better figure to use. But the difference is not that big 4.25 for current mean 4.77 if mean had increased in line with median. I'm not sure you can use the all earning figures because they are being pulled lower as more people are working part time.

ukcarper

that is not the point if we just used median the figure would be higher all of the time not just now

Graham_Devon

wotsthat wrote: »

Can I just check something? Do you understand that if you plot average full time male earnings OR all wages against house prices VS time both will be basically the same? I think you've spent a long time proving you don't. Honestly Graham - you haven't categorically proven that house prices are £40k overvalued.

The reason the graph is relevant is because good historical data is available and maintained from a reliable entity. It's not a conspiracy to try and make the earnings ratio look lower.

What graph would you trust - ONS all earnings vs. Land Reg? Or are we meant to 'just look out of the window'?

You clearly don't understand the point being made if you have wrote that.

ukcarper understands the point being made, Hamish understood the point being made, and I assume so too did ISTL.

Graham_Devon

ukcarper wrote: »

that is not the point if we just used median the figure would be higher all of the time not just now

We could go on and on. But we've achieved an agreement tonight, even if you are still massaging the figures to come up with 4.77!

We can't measure how to align mean and median and when to start, as we don't have the data, so we can't assume it's only 4.77...hence why I haven't stated how much they are overvalued, just made a point as to how the median and mean gap can increase, rendering the graph pretty much useless in reality.. I am assuming you have calculated that from a recent year...which is fine, as ultimately, it still proves the point I was making, but were talking about a 30 year graph, so you'd have to do it over 30 years....using a figure from half way through, or near the end of the period doesn't really work.

ukcarper

Graham_Devon wrote: »

We could go on and on. But we've achieved an agreement tonight, even if you are still massaging the figures to come up with 4.77!

We can't measure how to align mean and median and when to start, as we don't have the data, so we can't assume it's only 4.77. I am assuming you have calculated that from a recent year.

I've projected 1997 figures to 2011 all the figures we have in 1997 the ratio the figure was 3 it is now 4.77. If we were to use median the figure would be higher by the same percentage in 1997 and now as would be the long term average.

Graham_Devon

ukcarper wrote: »

I've projected 1997 figures to 2011 all the figures we have in 1997 the ratio the figure was 3 it is now 4.77. If we were to use median the figure would be higher by the same percentage in 1997 and now as would be the long term average.

Yer, which is pretty much what I did, so fair play. However, like I said, you've taken a figure half way through the graph and come up with a conclusion based on that.

As one set of data is covering 30 years, and the othe 15, the 4.77 is too low as 15 years data is missing. But again, like I say, we agree on the basis of the point made.

Probably easier all round to be honest to just say, using the median wage currently, house prices are 6x income and be done with it. Though I would like to point out if you use Hamish's preffered index, it's over 8x income

ukcarper

Graham_Devon wrote: »

Yer, which is pretty much what I did, so fair play. However, like I said, you've taken a figure half way through the graph and come up with a conclusion based on that.

As one set of data is covering 30 years, and the othe 15, the 4.77 is too low as 15 years data is missing. But again, like I say, we agree on the basis of the point made.

We don't know the figures before 1997 so we don't know the difference was greater or lower but it's the best we can do.

You should also remember that wage inflation has outstripped RPI by a considerable amount since 1997 let alone since 1982.

I believe Nationwide do use median and thier figures go back to 1953 and it is now 5x according to them.