A Goodbye To All That Buy To Let

[eric pebble]

On 30/09/2018 at 18:44, macca13 said:

Does not seem to make a difference in Kingston, loads of houses to let but Kingston university has built 2 new student accommodation blocks, this must be having an impact as in the past lots of the rentals were students.. same in ascot with Royal Holloway university increasing their student accommodation but numbers of students are down..

this must effect local landlords who bought to exploit students.. make sure they got in plenty of debt to make them rich..

will keep an eye on the to let signs see how long they stay empty.. 

SO much energy, time, resources etc etc. put into the very unproductive "business" of property speculation....  What a sad country this is....

[Ah-so]

A bit intense, but looks very interesting.

https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2018/determinants-of-distress-in-the-uk-owner-occupier-and-buy-to-let-mortgage-markets.pdf?la=en&hash=B1569BCB13CD86577927C22D166522BB51F556D9

Quote

Determinants of distress in the UK owner-occupier and buy-to-let mortgage markets

The mortgage market has played a central role in the global financial crisis. One particularly pressing question surrounds the conditions under which mortgage borrowers enter distress, ie get into arrears or default. This paper develops a novel micro dataset from residential mortgage loans which UK banks and building societies have pre-positioned with the Bank of England for use as collateral in exchange for central bank funding. The dataset is used to investigate the determinants of borrower distress as a function of borrower and loan-level stock/flow characteristics over the loans’ lifetime in the buy-to-let (BTL) and owner-occupier (OO) mortgage markets. We find systematic differences between these two markets, controlling for a range of loan and borrower characteristics as well as macro variables. Our main result shows that, adjusting for affordability, the loan-to-value ratio is reliably more important for borrower distress in the OO market than for distress in the BTL market, contradicting McCann’s (2014) results

 

[hotblack42]

38 minutes ago, Ah-so said:

A bit intense, but looks very interesting.

https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2018/determinants-of-distress-in-the-uk-owner-occupier-and-buy-to-let-mortgage-markets.pdf?la=en&hash=B1569BCB13CD86577927C22D166522BB51F556D9

 

Now this 'distress' thing is an interesting one.  If a house plummets in value but you still have your income then why does that create distress?
The only thing I can think of, is that some borrower-occupiers are structurally reliant on withdrawing equity for essential spending, but no one is that stupid..

I know, ridiculous right?  I mean what lender in their right mind would facilitate this?  .. Oh hang on.

[Ah-so]

2 hours ago, hotblack42 said:

Now this 'distress' thing is an interesting one.  If a house plummets in value but you still have your income then why does that create distress?
The only thing I can think of, is that some borrower-occupiers are structurally reliant on withdrawing equity for essential spending, but no one is that stupid..

I know, ridiculous right?  I mean what lender in their right mind would facilitate this?  .. Oh hang on.

Hence the double trigger observation and the reality that sharp house price falls tend to be accompanied by wider economic stress. 

[lastlaugh]

4 hours ago, Ah-so said:

A bit intense, but looks very interesting.

https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2018/determinants-of-distress-in-the-uk-owner-occupier-and-buy-to-let-mortgage-markets.pdf?la=en&hash=B1569BCB13CD86577927C22D166522BB51F556D9

 

I had a read of the paper.  I don’t agree with the conclusion - that BTLers are safer than OOers in the event of HPC - mainly because I think the assumptions are weak.  However if the BOE are generating excuses to further tighten the screw on our BTL scum, then it’s all good.

[Ah-so]

55 minutes ago, lastlaugh said:

I had a read of the paper.  I don’t agree with the conclusion - that BTLers are safer than OOers in the event of HPC - mainly because I think the assumptions are weak.  However if the BOE are generating excuses to further tighten the screw on our BTL scum, then it’s all good.

The paper does state that BTL borrowers have a higher distress rate than owner occupiers, but it has less to do with high LTVs for BTL mortgages. 

[spyguy]

4 minutes ago, Ah-so said:

The paper does state that BTL borrowers have a higher distress rate than owner occupiers, but it has less to do with high LTVs for BTL mortgages. 

LTV is meaningless gor btl.

Its yield you need to look at. Any btl with a gross yield less than 8% is on s hiding to nowhere.

[Neverwhere]

20 minutes ago, Ah-so said:

The paper does state that BTL borrowers have a higher distress rate than owner occupiers, but it has less to do with high LTVs for BTL mortgages. 

It seems that (over the relatively stable period they've looked at) they've confirmed the LTV-distress relationship at individual BTL loan level, but not at combined portfolio-and-main-residence level:

Quote

In constructing our borrower-level variables with the list above, we have made the following assumptions:

1. For each borrower-month combination, we have formed a total current LTV ratio that incorporates all loans of a given borrower at that time and divided it by the total current valuation of his/her properties.
2. If a borrower has at least one BTL loan, then that borrower is flagged as BTL in our BTL dummy.

[. . .]

Our main conjecture, contrary to McCann’s (2014) finding, is that the effect of the LTV ratio on distress would be lower for the BTL sector than for the OO mortgage market when assessed at the borrower level. This may seem counterintuitive as there is theoretical and empirical evidence that high LTV ratios, and in particular negative equity, may induce moral hazard (as an example, see the reasons set out in McCann, 2014). And this may be more so for investment properties than for an owner-occupied home, the latter of which a home-owner may want to avoid losing under all circumstances.

Much of this evidence is based on an analysis at the loan- or property-level. In fact, we can confirm this result in our loan-level robustness checks, as set out further below.

[. . .]

It should be noted that care should be taken when extrapolating these results to other time periods outside of our sample (or to other countries). In the UK, the period from 2013 to 2016 can be seen as being characterised by relatively low volatility (e.g. no financial crisis or recession in the UK, relatively low and stable interest rates, no general downturn in the housing market). We would therefore caution against generalising these results to periods of higher volatility

I'm not sure why they wouldn't conclude that the difference between the two might indicate an unwillingness or inability on the part of BTLers to even out financial difficulties across all of their properties, in contrast to the way that the authors have evened out LTVs across all of their properties?

[lastlaugh]

9 hours ago, Neverwhere said:

It seems that (over the relatively stable period they've looked at) they've confirmed the LTV-distress relationship at individual BTL loan level, but not at combined portfolio-and-main-residence level:

I'm not sure why they wouldn't conclude that the difference between the two might indicate an unwillingness or inability on the part of BTLers to even out financial difficulties across all of their properties, in contrast to the way that the authors have evened out LTVs across all of their properties?

I’m not really sure of the purpose of the paper.  Why was the paper commisioned?  Why is the BOE looking for evidence that the BTL sector can survive a modest HPC?

[Ah-so]

1 minute ago, lastlaugh said:

I’m not really sure of the purpose of the paper.  Why was the paper commisioned?  Why is the BOE looking for evidence that the BTL sector can survive a modest HPC?

Staff working papers are released weekly by the Bank - one of its roles is to be an economics research centre. 

I'm not sure that the paper does say that the BTL sector can survive a modest HPC. It is looking at a relatively short period since around 2012 when it has had access to a huge amount of loan level data on mortgages.

[spyguy]

56 minutes ago, lastlaugh said:

I’m not really sure of the purpose of the paper.  Why was the paper commisioned?  Why is the BOE looking for evidence that the BTL sector can survive a modest HPC?

Well .. what else would an economist do? Work ....

BoE is reposnbile for the steady operatin of the UK's banks.

Fluffed it a bit 2002-2008ish ....

Directly, or indirectly, they'll be making a case that BTL banks need to hold more capital.

Itsonly going to be a few years before any regulated epsoit taking bank is forced to sell off IO mortgages. The BoE wants rid.

They also want banks to hold more capital agianst commercial loans.

 

 

[lastlaugh]

2 hours ago, Ah-so said:

Staff working papers are released weekly by the Bank - one of its roles is to be an economics research centre. 

I'm not sure that the paper does say that the BTL sector can survive a modest HPC. It is looking at a relatively short period since around 2012 when it has had access to a huge amount of loan level data on mortgages.

Well, the BOE is a public body, using public funds.  Any research they do is to inform or support policy decisions.

Their conclusion is, and I quote:

“Our main result shows that, adjusting for affordability, the loan-to-value ratio is reliably more important for borrower distress in the OO market than for distress in the BTL market, contradicting McCann’s (2014) results.”

The conclusion contradicts previous research and the general theme of this thread.  

And I think the conclusion relies entirely on a false assumption, namely:

“Everything else equal, a BTL borrower is more likely to be able to withdraw equity from at least one of their properties, and use these funds to cross-subsidize other properties, if necessary.”

I don’t think the assumption stands up to scrutiny.

[lastlaugh]

1 hour ago, spyguy said:

Directly, or indirectly, they'll be making a case that BTL banks need to hold more capital.

But that paper, by painting a rosier picture of the robustness of the BTL market is making a case for the exact opposite.  Directly or indirectly.

[Ah-so]

1 hour ago, lastlaugh said:

Well, the BOE is a public body, using public funds.  Any research they do is to inform or support policy decisions.

Their conclusion is, and I quote:

“Our main result shows that, adjusting for affordability, the loan-to-value ratio is reliably more important for borrower distress in the OO market than for distress in the BTL market, contradicting McCann’s (2014) results.”

The conclusion contradicts previous research and the general theme of this thread.  

And I think the conclusion relies entirely on a false assumption, namely:

“Everything else equal, a BTL borrower is more likely to be able to withdraw equity from at least one of their properties, and use these funds to cross-subsidize other properties, if necessary.”

I don’t think the assumption stands up to scrutiny.

It is there to improve the understanding of macro-economics, which is relevant for public policy, but it it not prepared directly for specific policies. The BOE also does not use public funds - it is a profit making bank, that is wholly owned by HMT.

The analysis did find something interesting, i.e. that, "the loan-to-value ratio is reliably more important for borrower distress in the OO market than for distress in the BTL market", which is slightly unexpected, but I do not question its reliability. These positions might switch in a recesssion of course, but the period covered has been one of very low defaults.

I am not sure where you get the idea of a that "the conclusion relies entirely on a false assumption". I think that it is a statement of fact - BTL borrowers are more likely to remortgage that owner-occuperis. I do not know whether remortgages are sometimes used to cross-subsidize other properties, but it seems logical that a portfolio landlord would do so if they had surplus funds on one and a shortfall on another. It would be illogical not to.

Quote

Furthermore, we find that BTL borrowers are by about a third less sensitive to changes in the average LTV ratio. On the other hand, we find that BTL borrowers are by about 2.6% more likely to go into distress than OO borrowers.

We find that borrowers without BTL loans experience a higher distress probability to increases in their loan’s interest rates than BTL borrowers.

So while BTL borrowers are more likely to go into default than OO, they are less sensitive to LTV.

The final point - badly worded as it is, also suggests that OO are more likely to experience distress as interest rates rise. There could be plenty of reasons for this, and again, it is in a low interest rate pre-S24 environment. I do not, however, dispute the finding.

Edited October 23, 2018 by Ah-so

[iamnumerate]

22 hours ago, hotblack42 said:

Now this 'distress' thing is an interesting one.  If a house plummets in value but you still have your income then why does that create distress?
The only thing I can think of, is that some borrower-occupiers are structurally reliant on withdrawing equity for essential spending, but no one is that stupid..

I know, ridiculous right?  I mean what lender in their right mind would facilitate this?  .. Oh hang on.

If  you have to remortgage because you have a two year cheap fix, it will cause distress

[spyguy]

7 hours ago, lastlaugh said:

But that paper, by painting a rosier picture of the robustness of the BTL market is making a case for the exact opposite.  Directly or indirectly.

Well, they are looking for the wrong risks - OO is different to BTL.

Sure, io btl has ir risk, like oo. And the biggest ir risk btl faces is the need for banks to hold more capital for io btl.

But the biggest risk with rentals is voids. Something where theres just no comparison to residential OO.

BoE is daft trying to do a comparison. But they were overseeing RBS too, so theyve form on missing risk.

[Houdini]

15 minutes ago, spyguy said:

Well, they are looking for the wrong risks - OO is different to BTL.

Sure, io btl has ir risk, like oo. And the biggest ir risk btl faces is the need for banks to hold more capital for io btl.

But the biggest risk with rentals is voids. Something where theres just no comparison to residential OO.

BoE is daft trying to do a comparison. But they were overseeing RBS too, so theyve form on missing risk.

In reality the big issue for both OO and BTL is the same - it’s a sudden drop in income that triggers the distress.

For OO’s the people likely to have problems will be high LTV as chances are they wont  have savings to make up any income shortfalls, for BTLers it’s voids that are the issue but during the period the report covers I suspect it was easy to pay the interest either from savings or from other rental income. Whether they can still do that with S24 and banks being more careful with remortgaged is a different matter.

[Neverwhere]

6 hours ago, Ah-so said:

The analysis did find something interesting, i.e. that, "the loan-to-value ratio is reliably more important for borrower distress in the OO market than for distress in the BTL market", which is slightly unexpected, but I do not question its reliability. These positions might switch in a recesssion of course, but the period covered has been one of very low defaults.

What do you think might explain the difference in relationship between LTV and distress at BTL loan level and at BTL borrower level?

It seems like the method for assessing LTV at borrower level could conflate BTL borrowers who've experienced a slight drop in LTV across the board, BTL borrowers who've experienced a significant drop in LTV on a single rental property, and BTL borrowers who've experienced a significant drop in LTV on their own homes, all of whom might be expected to behave somewhat differently?

Quote

In constructing our borrower-level variables with the list above, we have made the following assumptions:

1. For each borrower-month combination, we have formed a total current LTV ratio that incorporates all loans of a given borrower at that time and divided it by the total current valuation of his/her properties.
2. If a borrower has at least one BTL loan, then that borrower is flagged as BTL in our BTL dummy.

 

[Neverwhere]

11 hours ago, lastlaugh said:

I’m not really sure of the purpose of the paper.  Why was the paper commisioned?  Why is the BOE looking for evidence that the BTL sector can survive a modest HPC?

I think that as Ah-so notes it's likely to improve their understanding of what's going on, rather than to pursue or support a specific policy.

As they've started to collect borrower level data, and the PRA have been asking lenders to assess borrowers' portfolios and not just the individual property they're lending against, it makes sense that they would be looking to see whether this information provides any new insight into the behaviour of a sector (BTL) which the Bank has previously expressed concern about.

[Bland Unsight]

Back in April 2016 higher rate of SDLT for additional properties was introduced. That was, of course, before the referendum. Here's what the prices of London flats have done since then, as per the Land Registry.

At £20k off a £430k peak any buy-to-let investor leveraged at, say, 60% LTV buying in July 2017  is now down £20k on £170k invested . More than 10% down on a fall of less than 5%. That's what leverage will do.

Edited December 19, 2018 by Bland Unsight

[Neverwhere]

On 25/05/2018 at 18:14, Bland Unsight said:

We're now looking back almost three years  - and how we've all grown over that time, perhaps none more so than the esteemed Bosher, whose dream of getting clause 24 all "cleared up" over the next few days has led to a transformation from a larval stage of quietly running a string of crappy terraced house somewhere in the Valleys into a beautiful butterfly - the nation's leading academic landlord.

Anyway, here's the latest data from the government on dwelling stock by tenure:

Source: Ministry of Housing, Communities and Local Government - Dwelling stock estimates: 2017, England (release date 24 May 2018, link).

If the PRS continues to shrink or if indeed the rate at which it shrinks increases things could get very interesting. Of course if house prices started to fall sharply and that contributed to investors selling up to capture gains before they vanished, or stop losses before they get monstered by them, then we could end up in a feedback loop of falling house prices calling forth highly motivated sellers and driving house prices down further. Even if all we see is a sell-off of the million homes added to the PRS since 2008, things could still get pretty exciting.

On 02/09/2018 at 11:10, Bland Unsight said:

Worth pointing out that the line that there are "4,000 rental homes are being sold off by landlords each month" as per the Which report is slightly careless way of saying what's going on. That's the net change in the PRS but other contributory flows include landlords selling to landlords (no net change), landlords purchasing additional buy-to-lets and owner occupiers becoming landlord by taking out buy-to-let mortgages on homes that were previously their residence. Hence it could be that there are 8,000 disposals and 4,000 additions, giving a net change of 4,000.

And of course, this is the entire PRS of which only about half is held on a buy-to-let mortgage.

On 17/10/2018 at 18:49, Neverwhere said:

125,000 down and counting...

Quote

Do measures that discourage buy-to-let investment increase rents? A Generation Rent paper

From its peak of 4.37m in Q2 2016, just at the point where the first measures began to take effect, the size of the private rented sector subsequently fell by 125,000 by Q2 2018 – a clear negative shock to the supply of private rented accommodation. But what happened to rents?

The ONS private rent index shows us that from April 2014 until February 2016 rents grew by 4.6%, adjusted for inflation, despite the buy-to-let rush ahead of the new stamp duty levy, raising the supply of rented properties in the short term.

But once the stamp duty surcharge was in place and the private rented sector began to shrink, real rents actually stopped rising, and broadly flatlined until January 2017. At this point, three months before restrictions on mortgage interest relief began to be phased in, real rents started falling again. As of August 2018, they have fallen by 3.2% in real terms.

Continuing falls in new BTL borrowing, potentially an early indication that the next dwelling stock estimates will show a further reduction in the size of the PRS?

Quote

Buy-to-Let Property Borrowing Is Losing Favor in the U.K.

Buy-to-let borrowing as a share of all mortgages has fallen to the lowest level in six years, according to data from the Bank of England. It had accounted for more than one fifth of such home loans at the start of 2016.

[Neverwhere]

Bearing in mind that there is a significant lag in these figures, as of March 2018 the PRS regained a little (less than a quarter) of its 2016/17 losses, and expansion of owner occupation continued on apace:

Dwelling stock estimates: 2018, England

[Switch625]

Looks like the vested interests at the BBC are beginning to panic about this. So how can they possibly turn the data into a "positive" HPI and BTL forever story?

The data they have used for this piece agrees with what you have identified, but the headline is ....

Soaring second home ownership hitting young people

https://www.bbc.co.uk/news/business-48637484

 

[Neverwhere]

19 minutes ago, Switch625 said:

Looks like the vested interests at the BBC are beginning to panic about this. So how can they possibly turn the data into a "positive" HPI and BTL forever story?

The data they have used for this piece agrees with what you have identified, but the headline is ....

Soaring second home ownership hitting young people

https://www.bbc.co.uk/news/business-48637484

I reckon that as the policy changes around BTL prevent sentiment from driving speculation beyond a certain price point, (and that price point is likely below where we are now given the tax burden on BTL properties is still increasing), so any pro-HPI element of a story about a post-2001 BTL boom has been largely rendered toothless at this point.

What then remains - and is actually helpful for a HPC - is the human interest element, which drives political sentiment about the negative impact of BTL on would be homeowners and especially younger voters.

I think this is a genuinely "unintended consequence"  of phasing in the BTL tax changes over several years: the government thought they were being very clever and slowly letting the steam out of the market without precipitating an out-and-out crash, but by tackling the problem so slowly they've created a situation where there is still an increasing amount of pressure for further policy measures against BTL before the last ones are anywhere close to working their way through the system.

[Neverwhere]

43 minutes ago, Switch625 said:

Looks like the vested interests at the BBC are beginning to panic about this. So how can they possibly turn the data into a "positive" HPI and BTL forever story?

On the subject of the general cultural bias at the BBC on housing and BTL, the new Economics Editor, starting - I think - later this month, had a very good chapter on BTL in his book The Default Line, so we might see a bit of better coverage on this in the future.