CML: FTB lending up 24% YoY

Graham_Devon

OK, but isn't a true measure of YOY the 12 monthly data points averaged out? You wouldn't fill in accounting or benefit forms stating that because you'd done 30 hours overtime, your yearly wage has now increased over the year by those 30 hours each month.

As I say, if a few hundred fewer people in December 2012 secured a mortgage than in December 2011, YOY lending would switch from 24% positive to around 10% negative.

That's some fluctuation, rendering the measure almost worthless.

Infact, I'm sure there are months in 2012 we could provide a working example.

Month on month is fine, but I'm struggling with your defintion of YOY, especially from an accounting perspective. It would be insane. You could have 11 months of dire accounts, down each and every month, and one good month where you've sold some property off as a one off, and claim profits are up 110% YOY.

the_flying_pig

Generali wrote: »

This is what I would describe as YoY, comparing the total lending for the month of November in 2012 with the total lending for the month of November in 2011.

That is also the measure most analysts would call YoY (year on year). MoM (month on month) would compare October 2012 with November 2012.

So if the number of mortgages in Dec 2011 was 20,000 and in Dec 2012 it turns out to be 18,000 then the number of mortgages will have fallen by 2,000 or 10% YoY and almost 20% MoM.

It would be possible to compare the total number of mortgages from the year to November 2011 with the total number for the year to November 2012 but it wouldn't serve the purpose of the stats which is to give a snapshot of the direction the market is moving in as a very approximate guide to the future.

If you want to look at the value of the loans rather than the number of loans (a more valid metric IMHO) then that is up a mere 23% YoY.

Disagree with all of this.

Imagine we were talking about sales data for HMV.

Would we compare Dec 11 sales with Dec 12 sales & use this information to conclude whether sales were up or down YoY?

Of course not.

We might well compare Dec 11 sales with Dec 12 sales & use this information to conlcude whether Christmas sales were up or down YoY.

But unless November as a month is of particular interest for some reason it's just ridiculous to use it in YoY calculations. You can't do it.

What might be possible, as per Haliwide indices, is to seasonally adjust it & use it to calculate MoM numbers.

Rinoa

The usual suspects are apoplectic.

They never contemplated the housing market improving.

They are disbelieving.

Graham_Devon

I've had to google it to make sure I wasn't going insane, and thankfully for me, it appears I'm not.

Heres how to work it out so that the data isn't as stated and wildly fluctuating on small numbers...

1. Subtract last year's number from this year's number. This gives you the total difference for the year. Hopefully, it's positive and indicates year-over-year growth, not loss.
2. Then, divide the difference by last year's number.
3. That gives you the year-over-year growth rate.

The OP is missing points 2 and 3.

Generali

the_flying_pig wrote: »

Disagree with all of this.

Imagine we were talking about sales data for HMV.

Would we compare Dec 11 sales with Dec 12 sales & use this information to conclude whether sales were up or down YoY?

Of course not.

We might well compare Dec 11 sales with Dec 12 sales & use this information to conlcude whether Christmas sales were up or down YoY.

But unless November as a month is of particular interest for some reason it's just ridiculous to use it in YoY calculations. You can't do it.

What might be possible, as per Haliwide indices, is to seasonally adjust it & use it to calculate MoM numbers.

I take your points and agree with them entirely but.....

...in my world (Investment Banks/Hedge Funds/Mutual Funds) this is a completely normal way to look at changes in data to use them to predict how things might change in the future. When I look at the amount of money lent in November 2011 and the amount of money lent in November 2012 that sends me a signal about what might happen to mortgage lending in December 2012, especially if a bunch of other data are telling me the same thing.

November is of particular interest as most Novembers are reasonably similar, more so than a November is with a June or a February. Novembers normally have similar weather to each other, also similar proximity to Xmas, number of days and public holidays, kids in school-ness and other things.

Seasonal adjustment works for me too in most times but this data set isn't seasonally adjusted AFAIK.

I don't think you're wrong it's just that I come at this from a different direction. Are you an accountant? I struggle with accountancy. I learn it and then when I wake up the next day there is a small pile of accounting on my pillow where it fell out of my ear overnight.

Generali

Graham_Devon wrote: »

I've had to google it to make sure I wasn't going insane, and thankfully for me, it appears I'm not.

Heres how to work it out so that the data isn't as stated and wildly fluctuating on small numbers...



The OP is missing points 2 and 3.

Errm, no it's not:

FTB lending up 24% YoY

That's the title of the thread. You'll probably see it in blue near the top of your screen.

the_flying_pig

Generali wrote: »

I take your points and agree with them entirely but.....

...in my world (Investment Banks/Hedge Funds/Mutual Funds) this is a completely normal way to look at changes in data to use them to predict how things might change in the future. When I look at the amount of money lent in November 2011 and the amount of money lent in November 2012 that sends me a signal about what might happen to mortgage lending in December 2012, especially if a bunch of other data are telling me the same thing.

November is of particular interest as most Novembers are reasonably similar, more so than a November is with a June or a February. Novembers normally have similar weather to each other, also similar proximity to Xmas, number of days and public holidays, kids in school-ness and other things.

Seasonal adjustment works for me too in most times but this data set isn't seasonally adjusted AFAIK.

I don't think you're wrong it's just that I come at this from a different direction. Are you an accountant? I struggle with accountancy. I learn it and then when I wake up the next day there is a small pile of accounting on my pillow where it fell out of my ear overnight.

sigh. i think that's taken the debate about as far as it can usefully go.

the 24% does not compare the amt of lending over the year to nov 12 with the amt of lending over the year to nov 11.

the 24% does compare the amt of lending in nov 12 with teh amt of lending in nov 11.

make of this 24% figure what you will.

wotsthat

Graham_Devon wrote: »

Month on month is fine, but I'm struggling with your defintion of YOY, especially from an accounting perspective. It would be insane. You could have 11 months of dire accounts, down each and every month, and one good month where you've sold some property off as a one off, and claim profits are up 110% YOY.

YoY is just a way of comparing two data points. If it was in a set of accounts, for example, I'd expect the management to qualify if the sales were unusually skewed to a particular month or were particularly volatile.

Nationwide, I think, use a three month average to compare YoY so if we were in an upwards trend (using the same data) their YoY would be a smaller number but bigger in a downwards trend compared to a 'proper' YoY.

As we're in an upward trend you could, if you had the inclination, calculate YoY using the rolling averages for monthly figures and arrive at a figure less than 24% YoY. I'd suggest that thinking about what the data is saying would be a better use of time than making an effort to do the calculations in a different way.

Generali

the_flying_pig wrote: »

sigh. i think that's taken the debate about as far as it can usefully go.

the 24% does not compare the amt of lending over the year to nov 12 with the amt of lending over the year to nov 11.

the 24% does compare the amt of lending in nov 12 with teh amt of lending in nov 11.

make of this 24% figure what you will.

I agree with you.

Can you be a little clearer as to why you think this is the wrong approach? I speak to the CIO at our largish fund management company every day and I'm sure he'd be interested in anything that stole a march over our rivals and our Mom and Pop clients would love to get a better return on their pension savings.

HAMISH_MCTAVISH

Rinoa wrote: »

The usual suspects are apoplectic.

They never contemplated the housing market improving.

They are disbelieving.

Nail, head, etc.