Equity Research www.sebenskilda.se Important. All disclosure information can be found on pages 28 – 30 of this document Unique data collaboration with Boreda AB Unique insights into the leverage of Swedish tenant-owner’s associations This report has been enabled by data provided by Boreda AB (www.boreda.se), a young company which provides research on Swedish tenant-owner’s associations. We have been allowed access to information on more or less all of Sweden’s tenant- owner associations which has improved our understanding of this segment of the Swedish housing market. Our conclusion after having analysed this data is that risks related to tenant-owner’s associations and their tenant-owners are underestimated. The perils of double leverage and poor transparency We argue that owners of apartments in Sweden are very often unaware of the debt in the tenant-owner’s associations they are part of. The rapid credit expansion among apartment owners, the short-term borrowing of the tenant-owner’s associations plus these associations’ very limited ability to absorb increased costs means that rather modest mortgage rate hikes (to around 7%) could lead to unpleasant surprises for significant numbers of apartment owners, which in turn could have an adverse effect on the price development of owner occupied apartments and bank stocks. However, interest rate expectations and margin trends prevent downgrades Interest rate expectations are currently trending downwards; should these expectations materialise, the risks discussed in this report will not materialise. On top of this, margins on residential mortgages in Sweden are expanding rapidly. Thus paradoxically, in the short run, being a mortgage lender could mean being in a sweet spot. This prevents us from making any ratings changes on the back of our findings. However, we urge anyone with interests in the Swedish banking sector to look into these risks if interest rate expectations change as they have done before. Banks Sweden Sector Comment 13 June 2011 Weakest link in Swedish housing market Average estimated fee increase in Swedish tenant-owner's associations if mortgage rates go to 7% -10% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 11,400 tenant-owner's associations sorted into 20 brackets by borrowing/sqm Tenant-owner's association fee increase -10% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 11,400 tenant-owner's associations sorted into 20 brackets by borrowing/sqm Tenant-owner's association fee increase Source: SEB Enskilda and Boreda Analysts Hampus Brodén +46 8 522 297 44 [email protected]Nicolas McBeath +46 8 522 298 06 [email protected]
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Equity Research
www.sebenskilda.se Important. All disclosure information can be found on pages 28 – 30 of this document
Unique data collaboration with Boreda AB Unique insights into the leverage of Swedish tenant-owner’s associations
This report has been enabled by data provided by Boreda AB (www.boreda.se), a young company which provides research on Swedish tenant-owner’s associations. We have been allowed access to information on more or less all of Sweden’s tenant-owner associations which has improved our understanding of this segment of the Swedish housing market. Our conclusion after having analysed this data is that risks related to tenant-owner’s associations and their tenant-owners are underestimated.
The perils of double leverage and poor transparency We argue that owners of apartments in Sweden are very often unaware of the debt in the tenant-owner’s associations they are part of. The rapid credit expansion among apartment owners, the short-term borrowing of the tenant-owner’s associations plus these associations’ very limited ability to absorb increased costs means that rather modest mortgage rate hikes (to around 7%) could lead to unpleasant surprises for significant numbers of apartment owners, which in turn could have an adverse effect on the price development of owner occupied apartments and bank stocks.
However, interest rate expectations and margin trends prevent downgrades Interest rate expectations are currently trending downwards; should these expectations materialise, the risks discussed in this report will not materialise. On top of this, margins on residential mortgages in Sweden are expanding rapidly. Thus paradoxically, in the short run, being a mortgage lender could mean being in a sweet spot. This prevents us from making any ratings changes on the back of our findings. However, we urge anyone with interests in the Swedish banking sector to look into these risks if interest rate expectations change as they have done before.
Banks
Sweden
Sector Comment
13 June 2011
Weakest link in Swedish housing market
Average estimated fee increase in Swedish tenant-owner's associations if mortgage rates go to 7%
Sector Comment Weakest link in Swedish housing market
2 13 June 2011
Contents Page
Background: stalling house prices....................................................................................3 No imminent threat of a house price drop.........................................................................3
The canary in the coal mine ...............................................................................................6 Flats and their tenant-owner’s associations ......................................................................6 TOA debt below radar in household statistics...................................................................7 Apartment owners tend to have high leverage..................................................................8 Rapid credit expansion......................................................................................................9 TOA debt as short-term as any household debt .............................................................10 The peril of double leverage............................................................................................10 The scale of the potential problem..................................................................................13 Two reasons this could become problematic ..................................................................18 One reason there is no problem short term ....................................................................19
Appendix I ..........................................................................................................................28 Target prices and risks .....................................................................................................29
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
3 13 June 2011
Background: stalling house prices No imminent threat of a house price drop First, let’s make one thing clear – we see no imminent threat to house prices in Sweden. We believe there are enough fundamental factors to largely justify house price trends of recent years (see Swedish banks and house prices - An in-depth look banks’ exposure to housing, 13 September 2010). The Swedish central bank reached similar conclusions based on various econometric and other approaches in a recent 400-plus page report.
However, we do seem to be at some kind of turning point at the moment where rising rates and tighter lending standards are starting to affect house prices.
Source: SEB Enskilda and Bank of International Settlements
The Swedish residential real estate price trend seems more or less uninterrupted (see chart above). However, a closer look indicates that apartment prices have shown a more negative trend than house prices in recent months (see next chart).
House and apartment price development on a regional basis – Sweden
90
100
110
120
130
140
150
160
170
180
190
Feb-
05
Jun-
05
Oct
-05
Feb-
06
Jun-
06
Oct
-06
Feb-
07
Jun-
07
Oct
-07
Feb-
08
Jun-
08
Oct
-08
Feb-
09
Jun-
09
Oct
-09
Feb-
10
Jun-
10
Oct
-10
Feb-
11
Flats Stockholm Flats Gothenburg Flats Malmoe
Flats medium sized cities Houses Stockholm Houses Gothenburg
Houses Malmoe Houses medium sized cities
Source: SEB Enskilda and Valueguard
Fundamentals justify house price trends for the most part
Turning point
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
4 13 June 2011
The dip in apartment prices is by no means yet a trend, but a couple of factors indicate somewhat weaker residential real estate prices from now on. Firstly, we have again reached a point where home owners are not getting paid to live in their houses, which should make home buyers more price sensitive. In the next chart, we have deflated the actual mortgage rate Swedes are paying with the growth in residential real estate prices. During most of the past 15 years, Swedish home owners have been getting paid to live in their houses, i.e. their monthly interest cost has been lower than the growth in the value of the house. The last time owning a home meant paying instead of making money was in late 2008. However, that situation did not last very long as the popular three-month mortgage rate in Sweden fell from almost 7% to 1.5% in just 12 months in connection with the financial crisis. This pushed house price growth up again and Swedes again started making money on their homes. That is less likely to happen this time around, at least based on money market expectations of future rates.
Volume-weighted Swedish residential mortgage rate deflated by house price growth, %
-10%
-8%
-6%
-4%
-2%
0%
2%
4%
6%
8%M
ar-9
7
Mar
-98
Mar
-99
Mar
-00
Mar
-01
Mar
-02
Mar
-03
Mar
-04
Mar
-05
Mar
-06
Mar
-07
Mar
-08
Mar
-09
Mar
-10
Mar
-11
Mortgage rate deflated with y-o-y residential real estate growth
Owning your home Makes you Money
Owning your home costs you Money
-10%
-8%
-6%
-4%
-2%
0%
2%
4%
6%
8%M
ar-9
7
Mar
-98
Mar
-99
Mar
-00
Mar
-01
Mar
-02
Mar
-03
Mar
-04
Mar
-05
Mar
-06
Mar
-07
Mar
-08
Mar
-09
Mar
-10
Mar
-11
Mortgage rate deflated with y-o-y residential real estate growth
Owning your home Makes you Money
Owning your home costs you Money
Source: SEB Enskilda and Datastream
As an additional indication, our house price indicator has stopped working. Historically, a model using short-term interest rates and consumer confidence has worked reasonably well in forecasting house price growth over a six-month horizon. However, this time around our model is not working very well. As the next chart shows, Swedish national house price growth has dipped earlier than implied by the trend in short-term rates and consumer confidence.
SEB Enskilda house price indicator and actual house price development
-15
-10
-5
0
5
10
15
Jul-9
3Ja
n-94
Jul-9
4Ja
n-95
Jul-9
5Ja
n-96
Jul-9
6Ja
n-97
Jul-9
7Ja
n-98
Jul-9
8Ja
n-99
Jul-9
9Ja
n-00
Jul-0
0Ja
n-01
Jul-0
1Ja
n-02
Jul-0
2Ja
n-03
Jul-0
3Ja
n-04
Jul-0
4Ja
n-05
Jul-0
5Ja
n-06
Jul-0
6Ja
n-07
Jul-0
7Ja
n-08
Jul-0
8Ja
n-09
Jul-0
9Ja
n-10
Jul-1
0Ja
n-11
Jul-1
1Ja
n-12
Jul-1
2Ja
n-13
Residential real estate prices, y/y ch %Forecasted house price growth based on recorded consumer confidence and interest rates, y/y ch %Forecasted house price growth based on forecasted consumer confidence and interest rates, y/y ch %
-15
-10
-5
0
5
10
15
Jul-9
3Ja
n-94
Jul-9
4Ja
n-95
Jul-9
5Ja
n-96
Jul-9
6Ja
n-97
Jul-9
7Ja
n-98
Jul-9
8Ja
n-99
Jul-9
9Ja
n-00
Jul-0
0Ja
n-01
Jul-0
1Ja
n-02
Jul-0
2Ja
n-03
Jul-0
3Ja
n-04
Jul-0
4Ja
n-05
Jul-0
5Ja
n-06
Jul-0
6Ja
n-07
Jul-0
7Ja
n-08
Jul-0
8Ja
n-09
Jul-0
9Ja
n-10
Jul-1
0Ja
n-11
Jul-1
1Ja
n-12
Jul-1
2Ja
n-13
Residential real estate prices, y/y ch %Forecasted house price growth based on recorded consumer confidence and interest rates, y/y ch %Forecasted house price growth based on forecasted consumer confidence and interest rates, y/y ch %
Source: SEB Enskilda
Indications of somewhat weaker residential real estate prices
Swedish national house price growth has dipped earlier than implied by the trend in short-term rates and consumer confidence
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
5 13 June 2011
House price growth stalling before our model implies could, in our view, be due to an emerging and vocal debate about a potential residential real estate bubble in Sweden. This has led to both voluntary and mandatory tightening of lending standards. In February 2010, the Swedish FSA announced that it wanted to cap residential mortgage loan LTVs. Since then, the banks have introduced even harsher maximum LTVs than the regulators and some of the banks have also increased cash flow requirements on mortgages for proposed borrowers. Meanwhile, suggestions of abolishing interest cost deductibility, higher mortgage risk weights and the reintroduction of real estate tax have been made by various stakeholders that would like to see the housing market cool down.
All in all, the data-points above could mean that house prices are stalling. Given the documented autocorrelation1 (i.e. when prices go up they continue to go up and vice versa) in the rate of house price changes, stalling prices could imply a negative house price trend for a period of time. The most recent real estate broker barometer published by SBAB shows that Swedish real estate brokers on aggregate now expect falling apartment values over the coming quarter (see chart below).
SBAB real estate broker barometer
Q105 Q106 Q107 Q108 Q109 Q110 Q111
Net, percent
Condominiums
Houses
Forecast Q211
A negative value indicates real estate brokers that believe in falling prices outnumber the ones who believe in increasing prices Source: SBAB
If falling prices were to become fact, the focus will shift more towards risk and banking investors will start looking for the weakest link in the housing and mortgage market. This is what we have focused on in this report.
1 See for example the Riksbank’s inquiry into the risks of the Swedish housing market (2011), pg. 27.
Stalling prices could imply a negative house price trend for a period of time
If prices fall, the search will be on for the weakest link in the housing and mortgage market
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
6 13 June 2011
The canary in the coal mine We believe the weakest links, or the ‘canary in the coal mine’, in the Swedish housing market are the tenant-owners (condominiums) and tenant-owner’s associations to which they belong. There are several reasons why:
Debts in tenant-owner’s associations are typically not reported as household debt in official statistics, which means these household credit volumes ‘fly below the radar’ in a sense. This means that levels of overall household debt in Sweden typically end up being underestimated.
Apartment prices are more volatile than house prices, which also has implications for the attractiveness of apartments as collateral (see House and apartment price development on a regional basis – Sweden chart at the start of this report).
Individuals’ loan-to-value ratios on apartments are typically slightly higher than for houses in Sweden.
Credit expansion among apartment owners has been very pronounced in recent years, which is cause for caution in our view. Whereas mortgage institutions’ lending with single family houses as collateral has increased by 2x or 9% per year over the past 10 years, lending to apartment buyers has risen by 7x or 21% per year. Part of this difference is because more and more buildings have been turned into tenant-owner’s associations, but even adjusted for this the expansion in credit has clearly been sizable.
Surprisingly, the interest fixing periods of tenant-owner’s associations are just as short term as they are on individual household debt, which means rate changes will feed through fast.
Buyers of tenant-owned apartments in Sweden are often very much unaware of their double-leveraged nature – close to 100% of owner-occupied apartments in Sweden are part of tenant-owner’s associations. This is a key cause for concern.
In terms of the risk of lending to apartment owners, banks’ stress tests are not sophisticated enough to capture potentially large risks relating to the financial situation of the tenant-owner’s associations to which the apartments belong.
Flats and their tenant-owner’s associations We have outlined below what it means to own an apartment that is part of a tenant-owner’s association (TOA). Anyone familiar with the ownership structure of Swedish owner-occupied apartments can go directly to the section TOA debt below radar in household statistics.
Tenant-owner’s associations – the basics explained A tenant-owner’s association is formed when a landlord approaches his/her tenants with an offer to sell them the apartments they are renting. Tenants then form a tenant-owner’s association, which will be the formal buyer of the building. Theoretically, the entire purchase price can be paid by the tenant-owner’s association and financed by a bank loan. Instead of paying rent to the landlord, the tenants would then be paying a fee to the tenant-owner’s association, which now needs to service its debt. Each individual apartment owner’s fee would then be proportionate to their share of the building’s total square metres. The most common practice is that part of the purchase price is paid by the tenant-owner’s association and financed with a bank loan and the remaining part of the purchase price is paid by the apartment holders directly to the seller of the building. Each apartment holder can then choose how much equity to insert into their apartment and take out a mortgage loan on the rest. This means there are typically two layers of debt in Swedish tenant-owned apartments: the individual’s own mortgage loan and then the debt of the tenant-owner’s association that needs to be serviced by the individuals living in the building. The interest costs of the tenant-owner’s association are financed by fees paid by the tenants to the association. These fees also need to cover any joint costs, e.g. maintenance, energy and garbage handling.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
7 13 June 2011
How tenants become home owners and tenant-owner's associations are formed
The chart above illustrates the basic transactions that take place when a tenant-owner’s association is formed. In the next chart we have constructed a hypothetical tenant-owner’s association to illustrate the double leverage that arises in the lion’s share of apartments in such associations.
Square meters 175 150 125 100 75 625Share of TOA debt, SEKm 4,2 3,6 3,0 2,4 1,8 15,0Share of TOA debt, SEKth/sqm 24,0 24,0 24,0 24,0 24,0 24,0
Hypothetical market value of flat, SEKth/sqm 60,0 60,0 60,0 60,0 60,0 60,0Market value of flat, SEKm 10,5 9 7,5 6 4,5 37,5Hypothetical LTV of individual household 75% 35% 0% 80% 90% 53%Mortgage loan of individual household, SEKm 7,9 3,2 0,0 4,8 4,1 19,9
Share of TOA debt + household's debt, SEKm 12,1 6,8 3,0 7,2 5,9 34,9Value of flat if zero leverage (incl TOA) 14,7 12,6 10,5 8,4 6,3 52,5Total loan to value ratio (LTV) 82% 54% 29% 86% 93% 66%
Source: SEB Enskilda
The chart above illustrates how flat owners in Sweden typically have two layers of leverage. We highlight this as a risk: a large proportion of tenant owners do not know how much indirect debt they have, how much of their monthly fee is made up of interest costs for that debt or, most importantly, what changes in interest rates could mean for their future fees. For example, if the owner of apartment no.3 is unaware of the finances of the tenant-owner’s association, they might feel unleveraged when in fact the underlying leverage is more like 30%.
In the example above, an existing building is transformed into a tenant-owner’s association. There are also examples where newly built buildings start life as tenant-owner’s associations. The only difference is that the builder has to market its apartments to people who become the tenant-owners as opposed to people already living in the house.
TOA debt below radar in household statistics Before looking at the potential problem of tenant-owner’s association debt from a bottom up approach, we would just like to make one general remark. Tenant-owner’s association debt is household debt. Despite this, it is not included in central bank statistics under household debt, but under “loans to non-financial corporations”. Therefore, to really capture Swedish household debt, these volumes need to be included. The left-hand chart below shows household debt to GDP for various European countries. In order to get a large comparison base, we have used end-2009 statistics (the debt/GDP ratios do not change much in a year). When we add the tenant-owner’s associations’ debt to total residential mortgage debt, Sweden stands out as one of the countries with the most residential mortgage debt. Also, the relatively low owner-occupation rate in Sweden indicates that the mortgage debt per home owner is actually among the highest in Europe.
Mortgage debt per home owner in Sweden is actually among the highest in Europe
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
Source: SEB Enskilda and European Mortgage Federation Source: SEB Enskilda and European Mortgage Federation
Apartment owners tend to have high leverage Despite the typical additional leverage in the tenant-owner’s association, apartment buyers on average have higher loan-to-value ratios on their apartments than single family home owners have on their houses. This is evident both from the breakdown of Nordea’s Swedish cover pool (mortgage loans that back up mortgage bond issuance) and new lending data compiled by the Swedish FSA last year.
Loan-to-value ratios in Swedish banks' cover pools, Q1 2010 Loan-to-value ratios on new Swedish mortgage lending
Source: SEB Enskilda and company data Source: SEB Enskilda and the Swedish FSA
It is not entirely clear why LTVs of apartment owners are higher than for single family house owners, but we believe it is to do with the higher turnover in the stock of apartments in Sweden compared with houses. In a market with rising residential real estate prices, borrowing on an individual object typically increases when they are turned over. The next chart illustrates the total value of sold tenant-owned apartments in Sweden in recent years. Only in the past six years, apartments worth around SEK 600bn have been sold. Considering that lending to tenant-owners is around SEK 400bn, it is clear that much of it has been added in transactions that have taken place in recent years.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
9 13 June 2011
Turnover in the market for tenant-owned apartments in Sweden
Turnove r tenant-owned flats, SEKbn (LHS) LTV on new lend ing to tenant-owners (RHS)
Source: SEB Enskilda, Statistics Sweden and the Swedish FSA
Higher new lending LTVs for apartments could be due to people taking the view that unexpected expenses are more likely for a house than an apartment. We believe this could be a misconception in many cases.
Rapid credit expansion Lending to apartment owners is one of the most rapidly growing segments of all in the Swedish banks’ balance sheets over the past 10 years. Total loan volumes are up 7x over the past 10 years, which translates into a compound annual growth rate of 21%. The rapid growth in lending to apartment buyers means the historical tendency for debt to be primarily in the tenant-owner’s associations’ balance sheets has turned into a situation where tenants have almost twice as much debt as the tenant-owner’s associations. This could be partly why many tenants are losing track of the debt in their respective associations, which we aim to show is the case.
Swedish mortgage institutions lending to tenant owners Combined debt of TOAs and their tenants (SEKm)
0
50 000
100 000
150 000
200 000
250 000
300 000
350 000
400 000
450 000
1995
-04
1996
-04
1997
-04
1998
-04
1999
-04
2000
-04
2001
-04
2002
-04
2003
-04
2004
-04
2005
-04
2006
-04
2007
-04
2008
-04
2009
-04
2010
-04
2011
-04
Swedish mortgage institutions' lending tenant-owners, SEKm
SEK 182bn or 43% of currently outstanding loans to tenant-owners have been written since early 2007
Loans to tenant-owners have increased 7x or 21% annually over the last 10 years
0
100 000
200 000
300 000
400 000
500 000
600 000
700 000
800 000
Jul-9
5
Jul-9
6
Jul-9
7
Jul-9
8
Jul-9
9
Jul-0
0
Jul-0
1
Jul-0
2
Jul-0
3
Jul-0
4
Jul-0
5
Jul-0
6
Jul-0
7
Jul-0
8
Jul-0
9
Jul-1
0
Lending to tenant-owner's associations, SEKm Lending to tenant-owners, SEKm
0
100 000
200 000
300 000
400 000
500 000
600 000
700 000
800 000
Jul-9
5
Jul-9
6
Jul-9
7
Jul-9
8
Jul-9
9
Jul-0
0
Jul-0
1
Jul-0
2
Jul-0
3
Jul-0
4
Jul-0
5
Jul-0
6
Jul-0
7
Jul-0
8
Jul-0
9
Jul-1
0
Lending to tenant-owner's associations, SEKm Lending to tenant-owners, SEKm
Source: SEB Enskilda and Statistics Sweden Source: SEB Enskilda and Statistics Sweden
Many tenants are losing track of the debt in their associations
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
10 13 June 2011
TOA debt as short-term as any household debt Swedish residential mortgage loans are relatively short term. Of the outstanding residential mortgage loans in Sweden today, roughly 70% mature within 12 months. When it comes to tenant-owner associations’ debt, the maturity profile seems more or less identical. The charts below illustrate the maturity profiles of mortgage institutes’ lending to households and non-financial corporations. Only 53% of total lending to non-financial corporations is tenant-owner associations’ debt. However, the boards that make the borrowing decisions of such associations are made up of household members. Given that the overall maturity structure of non-financial corporations is identical to that of households, we assume tenant-owner’s associations’ debt maturity profiles are very similar to those of individual households.
Mortgage institutions lending to households Mortgage institutions’ lending to non-financial corporations
> 2 years18%
1 year to 2 years13%
3 months to 1 year15%
<3 months54%
<3 months55%
>2 years22%
1 year to 2 years11%
3 months to 1 year12%
Source: SEB Enskilda and Statistics Sweden Source: SEB Enskilda and Statistics Sweden
The peril of double leverage As described above, not only are flats more leveraged than houses but there is typically additional leverage in the tenant-owner’s association. The big question is whether this is a problem. We argue that for the apartments where additional leverage in the tenant-owner’s association is high, this can get problematic.
To illustrate this, we have picked an existing central Stockholm tenant-owner’s association completely at random. The P&L and balance sheet of this association by the end of 2009 looked as follows (the 2010 annual report is not available yet). This is a fairly typical tenant-owner’s association with a total area of 1,719sqm divided into 31 flats. Three are rented to tenants by the tenant-owner’s association and 28 are owned by the tenants. From the P&L and balance sheet below, we have made a number of important conclusions.
The net result of the tenant-owner’s association is slightly negative, which means there is no room for lower income or higher costs.
The debt servicing costs are roughly equal to the total member fees so if interest costs double so must the member fees.
The interest cost as a percentage of long-term debt is 3.2%, which means that if mortgage rates go to 6-7% this tenant-owner’s association’s interest costs and member fees will in fact have to double.
The average duration of “long-term” debt is eight months, which means increased rates would feed through quickly.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
11 13 June 2011
Tenant-owner’s association (TOA) X in Stockholm, full-year 2009 annual report
P&L SEK SEK/Sqm CommentsIncomeMember fees 1 059 098 616 => Member fees/long term debt = 3,6%Other income 1 367 930 796
Long term debt 29 274 651 17 040 => Average maturity ~8 monthsShort term debt 498 859 290Reserves 0 0Total liabilies and equity 62 510 431 36 386
Source: SEB Enskilda in collaboration with Boreda AB
Having been in the market for flats in Stockholm ourselves, we believe apartment buyers in Sweden generally have only a very vague understanding of tenant-owner’s associations’ P&Ls and balance sheets. This includes a lack of awareness of tenant-owner’s association indebtedness. Without a good understanding of this, you would not have visibility on the future development in member fees either. This is presumably why the majority of flat owners/buyers seem to view tenant-owner’s association fees as a constant variable. Real estate brokers also often state “no planned increase of member fees” when they are marketing flats.
We went to the building of the tenant-owner’s association described above and collected the names of the tenants. We were able to obtain the phone numbers for 19 of the 28 tenants that owned their own flats and 13 spoke to us about their association’s finances. The 13 included the chairman, the wife of the ex-chairman and the treasurer of the tenant-owner’s association. These three individuals knew roughly how much debt the tenant-owner’s association had but only the current chairman had a good understanding of what rate increases would mean for member fees, This means the current treasurer was not even sure. The remaining 10 individuals had absolutely no idea of how much debt their tenant-owner’s association had and thus no view on their own share of that debt or on the effect on member fees from changes in interest rates. Keep in mind that this tenant-owner’s association was picked at random. It is relatively representative when it comes to the duration of long-term debt. It does however have above-average tenant-owner’s association debt per sqm, which makes the lack of knowledge about debt among the majority of the tenant-owners in the building all the more alarming. As we will show below, there are many tenant-owner’s associations in Sweden with equally high debt per sqm.
How nasty could a negative surprise get for unaware tenants in the example TOA? Most tenant-owners in the example TOA were unaware of their level of indirect debt. The question is thus whether rising rates could indicate a problem for Swedish apartment owners. It depends on the interest rate scenario. If we assume the short-term residential mortgage rate goes to 7%, this is roughly 2pp higher than is currently expected by the money market for end-2013, but by no means an extreme worst case scenario. The new P&L would look as shown in the next chart. To prevent the tenant-owner’s association from making a loss, it needs to increase annual member fees by 106% to SEK 1,269/sqm.
We believe flat buyers in Sweden in general have only a very vague understanding of tenant-owner’s associations’ P&Ls and balance sheets
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
12 13 June 2011
Tenant-owner’s association (TOA) X in Stockholm, after rate increase from 3.2% to 7%
P&L SEK SEK/Sqm CommentsIncomeMember fees 2 181 418 1 269 => Member fees/long term debt = 7,5%Other income 1 367 930 796
Long term debt 29 274 651 17 027Short term debt 498 859 290Reserves 0 0Total liabilies and equity 62 510 431 36 358
Source: SEB Enskilda and Boreda AB
What could this mean for a typical flat owner in the building? The next example is based on a hypothetical tenant who has paid SEK 55,000 per sqm (the average in central Stockholm) for a 150sqm apartment and that 70% (the average among Swedish apartment buyers in recent years) of the purchase is debt financed.
Negative surprise potential when rates go up from 3.2% to 7% for above TOA
2009
2010 Change
(%)Change
(SEK 000s) Square meters 150 150 Price, SEK 000s/sqm 55 55 Market value of flat, SEK 000s 8,250 8,250 Loan to value ratio on individual flat 70 70 Loan on individual flat, SEK 000s 5,775 5,775
Rate sensitivity if member fee had actually been constant Interest cost on household's loan, SEK 000s 128 283 121 155Unchanged member fee, SEK 000s 92 92 0 0Total cost, SEK 000s 220 375 70 155
Negative surprise SEK 000s 98 as % of total cost increase (%) 39 Source: SEB Enskilda
For this hypothetical tenant, the increase in total costs (member fee plus own debt servicing cost) when rates go from 3.2% to 7% is 115% or SEK 253,000 per year. If the same tenant thought the member fee would be a constant instead of moving in line with rates, SEK 98,000 or close to 40% of the annual SEK 253,000 cost increase would come as a surprise. See the next chart for foreseen and unforeseen cost increases in different mortgage rate scenarios.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
13 13 June 2011
Foreseen and unforeseen cost increase in different mortgage rate scenarios
0%
20%
40%
60%
80%
100%
120%
140%
160%
4% 5% 6% 7% 8%
Mortgage rate level
Tota
l inc
reas
e in
sum
of d
ebt s
ervi
cing
cos
t and
TO
A
fee
Foreseen cost increase Unforeseen cost increase
Source: SEB Enskilda
To illustrate the potential problem with unnoticed debt in tenant-owner’s associations, consider the following. Imagine an individual buying a share in a tenant-owner’s association for SEK 55,000 per sqm. If the individual’s loan-to-value ratio is 70%, his or her debt will be SEK 38,500 per sqm. An additional SEK 17,000 in debt per sqm through the TOA then is no small addition. Also, consider the following – the interest cost on the individual’s debt is 30% tax deductible whereas the interest cost on the tenant-owner’s association’s debt is not tax deductible at all. This means that in a scenario where mortgage rates are, e.g. 5%, the net debt servicing cost per sqm for the individual’s direct borrowing is SEK 1,080 per year (SEK 55,000 x 4% x 70%). The net debt servicing cost for the indirect debt at the same time would be SEK 680 per year (SEK 17,000 x 4%). This means that the indirect debt servicing cost net of taxes is almost 40% of total direct and indirect debt servicing costs, even though the indirect debt is “only” around 30% of the sum of direct and indirect debt.
Given this example, it should come as no surprise that we disapprove of banks tending to stress test only the tenant-owners’ direct debt for rate increases and not indirect debt (i.e. their share of the TOA’s debt) when they consider new lending. Stress tests that keep the TOA fee constant regardless of the interest rate scenario do not capture the real sensitivity. There will be individuals that pass these tests that would not if the fee increases in the same interest rate scenario were taken into account. For example, some banks stress test home buyers for a mortgage rate of 9%. Based on the example above, the cost increase that would show up in the bank’s stress test would be SEK 236,000 whereas the real cost increase would be SEK 385,000. This is no small figure considering for example that the median annual household income after tax in Stockholm is around SEK 490,000.
The scale of the potential problem To get a sense of the scale of the potential problem, we have drawn on Boreda’s unique database on the financials of Swedish tenant-owner’s associations. It is based on the most recently released P&Ls and balance sheets and captures 11,400 tenant-owner’s associations responsible for roughly 85% of borrowing by such entities in Sweden. Each bracket includes 570 tenant-owner’s associations. For the full dataset, see appendix I. This data enables us to make a number of important conclusions.
We disapprove of banks tending to stress test only the tenant-owners’ direct debt for rate increases and not indirect debt (their share of the TOA’s debt) when considering new lending
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
14 13 June 2011
Share of total Swedish TOA borrowing per bracket (570 TOAs per bracket)
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Perc
ent o
f tot
al o
utst
andi
ng b
orro
win
g by
Sw
edis
h te
nant
-ow
ner's
asso
ciat
ions
Source: SEB Enskilda and Boreda
The average Swedish TOA does not look very problematic Our first conclusion is that for the average tenant-owner’s association, the problem is not a big one. For the average tenant-owner’s association, the debt per sqm is SEK 4,811 (not SEK 17,000 as in our example TOA above). If one had bought an apartment for SEK 55,000 per sqm (the average in Stockholm), the debt servicing cost net of taxes “hidden” in the member fee is 15% of total direct and indirect debt servicing costs. An unforeseen cost increase in the event of interest rate rises should not therefore be a particularly big problem. As is often the case when it comes to credit risk, one needs to look at the tails rather than just the averages.
Seems reasonable to take a closer look at one-fifth of the lending We believe that perhaps 10% of the TOAs are at risk of causing their tenant-owners problematic surprises when rates go up. These are the tenants in the tenant-owner’s associations that have more than SEK 10,000 of debt per sqm. These TOAs make up 23% of outstanding loans to tenant-owner’s associations and make up 10% of the total number of associations (see share of total Swedish TOA borrowing per bracket in the next chart).
When tenant-owner’s association debt gets to SEK 10,000/sqm or above, the share of “hidden” total net debt servicing costs for the tenant-owner approaches or surpasses 30% for anyone who has bought the apartment for SEK 55,000 per square meter with 70% loan financing. We have highlighted these brackets in the following chart. We have applied Boreda’s data to the preceding charts to shed further light on the risks facing tenant-owners in those TOAs.
We believe that perhaps 10% of the TOAs are at risk of causing their tenant-owners problematic surprises when rates rise
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
15 13 June 2011
Share of tenant-owner's association debt in various debt/sqm brackets
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Tena
nt-o
wne
r's a
ssoc
iatio
n de
bt/s
qm, S
EK
Each bar represents 5% of the total
number of tenant-owner's associations
Source: SEB Enskilda and Boreda
Unsurprisingly, interest costs make up a large share of costs for indebted TOAs Using Boreda’s database, we can look into the P&L structure for each of these brackets. In the next chart, the costs of the average tenant-owner’s association are broken down into debt servicing costs and other costs (such as maintenance, garbage handling, energy) per sqm.
As illustrated in the next chart, the TOA debt servicing cost as a share of total costs, and thus as a share of total member fee, is high in the most indebted tenant-owner’s associations.
Breakdown of costs in tenant-owners associations sorted by indebtedness
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Ann
ual c
ost p
er s
quar
e m
eter
, SEK
Average non-interest cost/sqm Average interest cost/sqm
Source: SEB Enskilda and Boreda
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
16 13 June 2011
Most indebted TOAs don’t generate surpluses and so can’t absorb rate increases The next chart also makes it clear that in the two most indebted brackets, the average net result of the tenant-owner’s associations is zero or negative. This means any increase in debt servicing costs would have to be matched by a similar increase in member fees.
Tenant-owner's association result as a percentage of long-term debt
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Tena
nt-o
wne
r's a
ssoc
iatio
ns' n
et re
sult/
long
term
deb
t
Source: SEB Enskilda and Boreda
Interest cost increase equal to fee increase in TOAs that make up 11% of borrowing The next chart illustrates member fees and debt servicing costs as a percentage of long-term debt. It shows that in the last bracket (11% of lending to tenant-owner’s associations), debt servicing costs are more or less identical to member fees. As in our example TOA above, this means a percentage increase in interest costs necessarily leads to the same percentage increase in member fees.
Tenant-owner's association interest costs and member fee as percent of long term debt
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Average interest cost/Long term debt Average member fee/Long term debt
Source: SEB Enskilda and Boreda
It therefore seems that maybe 10% of lending is to TOAs where the financial position is very similar to the one we used as an example earlier in this report.
Based on the data above, we can analyse the sensitivity of TOA member fees to interest rate changes. We have simply assumed the TOAs need to keep their results around zero and thus need to compensate for any deficits created by rate hikes by increasing fees.
Any increase in debt servicing costs would have to be matched by a similar increase in member fees
A percentage increase in interest costs necessarily leads to the same percentage increase in member fees
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
17 13 June 2011
Assuming an arbitrary mortgage rate level of 7%, the picture we get is illustrated in the left-hand chart below.
TOA member fee change necessary if mortgage rates go to 7% Interest rate sensitivity of TOA fee in most indebted bracket
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Tena
nt-o
wne
r's a
ssoc
iatio
n fe
e in
crea
se
0%
20%
40%
60%
80%
100%
120%
140%
4% 5% 6% 7% 8% 9%
Hypothetical mortgage rate
Cha
nge
in T
OA
mem
ber f
ee a
s ef
fect
of r
ate
chan
ge
TOA member fee change in most indebted bracket at various interest rate levels
Source: SEB Enskilda and Boreda Source: SEB Enskilda and Boreda
There is one bracket that clearly stands out in the left-hand chart above. For the most indebted bracket, the tenant-owner’s associations’ fee would on average need to almost double if mortgage rates were to go to 7%. The interest rate sensitivity in this particular bracket is illustrated in the right-hand chart above. From this chart, it is plain that stress tests based on an 8-9% mortgage rate but assuming the TOA fee is constant do not capture the full picture. It remains to be seen what will happen when Swedish apartment buyers realise that TOA fees are not constant, but could fluctuate fairly significantly with interest rates. Even after such a realisation, many people will still have great difficulty in assessing the risk in the P&Ls and balance sheets of TOAs. People are generally just not fluent in accounting. We believe the reaction among ordinary apartment buyers could be to just apply a general risk premium across the board, which in such a case could affect apartment prices in general relative to house prices. In such a case, there is a risk that the prices of all apartments, not only those in highly leveraged tenant-owner’s associations, will be negatively affected.
Another way of illustrating how residential mortgage rates of around 7%, let alone 8-9%, could be problematic is to take the aggregated debt of tenant-owners and tenant-owner’s associations in Sweden and multiply it by the average mortgage rate that Swedes pay. This should be a good proxy of the aggregated debt servicing cost of TOAs and their tenant-owners in Sweden. At the average mortgage rate of 3.7% in Sweden in April, the annual debt servicing cost would be SEK 21bn – not far from the all-time high of SEK 24bn by end-2008. If rates were to go to 7%, the total debt servicing cost would rise to an estimated SEK 39bn, well above the current all-time high.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
18 13 June 2011
Estimated TOAs’ and tenant-owners' after tax debt servicing cost, SEKm
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
50,000
Mar
-96
Mar
-97
Mar
-98
Mar
-99
Mar
-00
Mar
-01
Mar
-02
Mar
-03
Mar
-04
Mar
-05
Mar
-06
Mar
-07
Mar
-08
Mar
-09
Mar
-10
Mar
-11
TOA debt servicing cost, SEKm TO debt servicing cost, SEKm
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 9%
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 8%
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 7%
CoMbined debt servicing cost of Swedish tenant-owner's associations and tenant owners today
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
40,000
45,000
50,000
Mar
-96
Mar
-97
Mar
-98
Mar
-99
Mar
-00
Mar
-01
Mar
-02
Mar
-03
Mar
-04
Mar
-05
Mar
-06
Mar
-07
Mar
-08
Mar
-09
Mar
-10
Mar
-11
TOA debt servicing cost, SEKm TO debt servicing cost, SEKm
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 9%
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 8%
Combined debt servicing cost of Swedish tenant-owner's associations and tenant owners if rates go to 7%
CoMbined debt servicing cost of Swedish tenant-owner's associations and tenant owners today
Source: SEB Enskilda and Statistics Sweden
Two reasons this could become problematic Swedish banks’ lending to Swedish tenant-owners and tenant-owner’s associations is between 5% (Nordea) and 15% (Handelsbanken) of total lending so exposures are significant (see chart below).
Break down of Swedish banks’ lending Q1 2011
0%
20%
40%
60%
80%
100%
Handelsbanken SEB Swedbank Nordea
Corporate, municipal and other lending CEE residential mortgage lendingOther Nordic residential mortgage lending Swedish single family housesSwedish tenant-owner associations Swedish tenant-owners
Source: SEB Enskilda and company data
If Nordea Sweden is representative of the whole market, the loan-to-value ratio on outstanding tenant-owner loans should be 55-60%. On the face of it, this does not seem too intimidating. However, keep in mind that apartment prices fell more than house prices in the rate increase-related dip in house prices during 2008. Apartment prices fell by almost 20% in Sweden (and even more than that in Stockholm and Malmö) from late 2007 to late 2008 when rates climbed. In light of this and the 33% price increase in apartment prices since the trough until today, a 30% drop in apartment prices is not an outrageous scenario if mortgage rates were to go to 7%. If prices fall by 30%, a 55% average loan-to-value ratio would turn into a 79% average loan-to-value ratio. That would probably be enough to turn the share price momentum of large Swedish mortgage lenders like Handelsbanken and Swedbank quite negative. A price fall like that would include lots of mortgage loans going above loan-to-value ratio of 75%, which would disqualify them from inclusion in the cover pools that back covered bond issuance.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
19 13 June 2011
In a climate with falling prices and shrinking cover pools, it is not inconceivable that mortgage bond buyers demand better spreads. Higher funding costs would then need to be passed on to borrowers, which would cause house and apartment prices to come down further, this could affect consumer confidence and spending which in turn could affect employment which would mean house prices could take another hit. Thus, even though it would take quite a severe interest rate scenario for loan losses from these segments to threaten Swedish banks’ capital positions, it is not too difficult to envisage scenarios that could hurt banks’ share prices. Still, we believe the main risk to credit quality is an indirect one. This is because falling house prices typically imply a decline in consumer spending, which could in turn mean worsening corporate credit quality. The charts below illustrate the historical correlation between house prices and consumer spending on the one hand (left-hand chart) and consumer spending and corporate bankruptcies on the other (right-hand chart). This is the second reason why falling house prices, despite decent residential mortgage credit quality, could be bad news for the momentum in the share prices of banks.
Growth in residential real estate prices and consumer spending Growth in consumer spending and corporate bankruptcies
-4%
-2%
0%
2%
4%
6%
8%
10%
12%
14%
16%
Q11996
Q11997
Q11998
Q11999
Q12000
Q12001
Q12002
Q12003
Q12004
Q12005
Q12006
Q12007
Q12008
Q12009
Q12010
Q12011
Res
iden
tial r
eal e
stat
e pr
ices
, y-o
-y g
row
th
-1%
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%C
onsu
mer
spe
ndin
g, y
-o-y
gro
wth
Re s id e ntial re al e s tate p rice , y -o -y g ro wth Co nsum e r sp e nd ing , y -o -y g ro wth
-60%
-40%
-20%
0%
20%
40%
60%
80%
100%
Q11996
Q11997
Q11998
Q11999
Q12000
Q12001
Q12002
Q12003
Q12004
Q12005
Q12006
Q12007
Q12008
Q12009
Q12010
Q12011
Cor
pora
te b
ankr
uptc
ies,
y-o
-y g
row
th (r
ever
sed
axis
)
-1%
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
Con
sum
er s
pend
ing,
y-o
-y g
row
th
Residential real estate price, y-o-y growth Consumer spending, y-o-y growth
Source: SEB Enskilda and Datastream Source: SEB Enskilda and Datastream
One reason there is no problem short term We are negatively surprised that our analysis has led us to the conclusion that residential mortgage rates of a mere 7% could become quite problematic for apartment holders in Sweden. After all, mortgage rates were close to that level as recently as late 2008 and that was before regulators started demanding that banks fund themselves longer term, which is more expensive (i.e. there are more funding costs that need to be passed on to clients). See historical residential mortgage rates below.
Residential mortgage rates of state owned mortgage lender SBAB
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
1-10
-199
7
7-10
-199
7
1-10
-199
8
7-10
-199
8
1-10
-199
9
7-10
-199
9
1-10
-200
0
7-10
-200
0
1-10
-200
1
7-10
-200
1
1-10
-200
2
7-10
-200
2
1-10
-200
3
7-10
-200
3
1-10
-200
4
7-10
-200
4
1-10
-200
5
7-10
-200
5
1-10
-200
6
7-10
-200
6
1-10
-200
7
7-10
-200
7
1-10
-200
8
7-10
-200
8
1-10
-200
9
7-10
-200
9
1-10
-201
0
7-10
-201
0
1-10
-201
1
3 months 1 years 2 years 3 years 4 years 5 years 7 years 10 years
Source: SEB Enskilda and company data
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
20 13 June 2011
If money markets were to start implying mortgage rate hikes towards 7%, it would be very difficult to have anything but negative ratings on the large mortgage lenders in Sweden, in our view. However, at the moment there is very little to indicate that we are heading towards that level in the short to medium term. Interest rate expectations are even falling at the moment. The left-hand chart below shows money market expectations on three-month Stibor; the right-hand chart illustrates the change in those expectations.
Forward rate agreements on three-month Stibor Change in forward rate agreement rate for 3M Stibor end-2012E
3.343.323.293.263.213.17
3.113.04
2.95
2.85
2.69
2.45
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
3.6
Q31
1
Q41
1
Q11
2
Q21
2
Q31
2
Q41
2
Q11
3
Q21
3
Q31
3
Q41
3
Q11
4
Q21
4
3M STIBOR FORWARD RATE AGREEMENTS, %
1.5
2
2.5
3
3.5
4
14/0
9/20
09
14/1
1/20
09
14/0
1/20
10
14/0
3/20
10
14/0
5/20
10
14/0
7/20
10
14/0
9/20
10
14/1
1/20
10
14/0
1/20
11
14/0
3/20
11
14/0
5/20
11
Ex pected lev el of 3M STIBOR in December 2012
Source: SEB Enskilda and SIX Source: SEB Enskilda and SIX
If we use three-month Stibor expectations from the money market and add to the expected future rate the current difference between three-month Stibor and the three-month residential mortgage rate at Swedish banks, we reach an implied expectation of a three-month residential mortgage rate. On this measure, the mortgage rate is currently expected to be 4.58% by end-2013. The right-hand chart below shows that in such an interest rate environment, the least indebted TOAs can even lower fees (as they are generating surpluses today). In this rather modest interest rate hike base-case scenario, the most indebted 5% of tenant-owner’s associations would see their TOA fees rise by one-third.
Three-month mortgage rate implied by FRAs TOA member fee change if mortgage rates go to 4.58%
Tenant-owner's associations sorted into 20 brackets by borrowing/sqm
Tena
nt-o
wne
r's a
ssoc
iatio
n fe
e in
crea
se
Source: SEB Enskilda and SIX Source: SEB Enskilda and Boreda
Mortgage lending might even be a sweet spot short term When the inherent risk for tenant-owners in Sweden dawned on us while working on this report, we carefully contemplated what it meant for the main mortgage lenders in Sweden (Handelsbanken and Swedbank). However, rate expectations are on their way down and current expectations mean only 5% of tenant-owner’s associations have to hike fees by a mere third and the remaining 95% only marginally over the coming three years. Under such circumstances, it is difficult to see any near-term need to take a negative stance on the shares. Therefore, although we are slightly alarmed that the safety margin is not bigger, we also conclude that we do not seem to be heading towards a dangerous interest rate level in the foreseeable future. In fact, short term, being a large mortgage lender in Sweden looks like being in a sweet spot. This is because pricing discipline among the main players is better than it has been for a very long time.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
21 13 June 2011
If we subtract the current yields on Swedish banks’ mortgage bonds from their current list prices for residential mortgages, this should be a decent proxy of where new lending margins are heading.
Implied new lending margin 3M residential mortgages, % Implied new lending margin five-year residential mortgages, %
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
10/0
6/19
97
10/0
6/19
98
10/0
6/19
99
10/0
6/20
00
10/0
6/20
01
10/0
6/20
02
10/0
6/20
03
10/0
6/20
04
10/0
6/20
05
10/0
6/20
06
10/0
6/20
07
10/0
6/20
08
10/0
6/20
09
10/0
6/20
10
10/0
6/20
113 month mortgage margin (rolling 30 days)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
10/0
6/19
97
10/0
6/19
98
10/0
6/19
99
10/0
6/20
00
10/0
6/20
01
10/0
6/20
02
10/0
6/20
03
10/0
6/20
04
10/0
6/20
05
10/0
6/20
06
10/0
6/20
07
10/0
6/20
08
10/0
6/20
09
10/0
6/20
10
10/0
6/20
11
5 year mortgage margin (rolling 30 days)
Source: SEB Enskilda, SBAB and Datastream Source: SEB Enskilda, SBAB and Datastream
As seen above, the new lending margins are increasing almost explosively at the moment, heading towards 85bp on three-month loans and 125bp on five-year loans. Some of the banks are coming from back book mortgage margins of 55-60bp. Should discipline remain, it would mean very strong residential mortgage net interest income growth. After all, we are talking new lending margins several tens of basis points above back book margins. We estimate total Swedish residential mortgage lending net interest income of around SEK 10bn. This would increase by roughly SEK 1.5bn for every 10bp that the total residential mortgage book margin expands. For the two largest mortgage lenders in Sweden, Swedbank and Handelsbanken, this translates into a 4-5% positive effect on pre-tax profit for every 10bp in mortgage margin expansion. See the summary of sensitivity to Swedish residential mortgage margin changes below.
Effect on NII and operating profit from a 10bp change in Swedish mortgage margin
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
6,0%
Swedbank SHB SEB Nordea
Effect on NII as % of NII Effect on NII as % of operating profit
Source: SEB Enskilda
The above very positive trends seem highly likely to play out before any interest rate scenario that would cause the risks we have discussed in this report to materialise. As long as that is the case, we see no need to downgrade Swedbank and Handelsbanken. However, anyone that shares our concern for the long term and would rather avoid these risks completely, be aware that our top pick, Nordea, has the smallest exposure to these potential risks as well as the biggest upside to our target.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
22 13 June 2011
For our full investment case on Nordea, see previous reports (e.g. Strong report, but of slightly low quality, 29 April). In short, we like Nordea for the following reasons:
Nordea has significantly underperformed other Nordic banks and for a very long time.
The poor absolute and relative performance has occurred alongside the most positive earnings revisions trend by far for all the Swedish banks.
The shares trade below 9x 2012E earnings.
The implied long-term return on tangible equity is low historically compared with both other Swedish banks and the general market.
There is significant upside of around 40% to our target price.
Our estimates of the upside/downside to target prices and our forecasts versus consensus are outlined in the following table.
Nordic banks ranking
Recommendation Price target (SEK) Up-/downside 1 month EPS estimate revision (%) Current Previous Current Previous target price (%) 2011E 2012E 2013E Nordea Bank Buy Hold 98 98 43 0 0 0DnBNOR Buy Hold 106 106 36 0 0 0Swedbank Buy Buy 135 135 32 0 0 0Jyske Bank Buy Buy 294 294 30 0 0 0Pohjola Bank Buy Hold 11 11.1 27 0 0 0Handelsbanken Buy Hold 246 246 26 0 0 0Sydbank Hold Hold 150 150 23 0 0 0Danske Bank Hold Hold 120 120 20 0 0 0SEB Unrated Unrated Unrated Unrated Unrated 0 0 0 Average 30 0 0 0 Recommendation Price target (SEK) Share SEB Enskilda vs consensus EPS (%) SEB Enskilda Consensus SEB Enskilda Consensus price 2011E 2012E 2013E Nordea Bank Buy Buy 98 85 68 -3 3 2DnBNOR Buy Buy 106 96 78 6 2 4Swedbank Buy Buy 135 125 103 6 12 8Jyske Bank Buy Buy 294 290 226 7 1 0Pohjola Bank Buy Buy 11 11 9 4 0 -6Handelsbanken Buy Hold 246 242 195 1 0 0Sydbank Hold Buy 150 165 122 0 -8 3Danske Bank Hold Buy 120 136 100 -4 -12 -8SEB Unrated Hold Unrated 60 51 4 -4 -1 Average 2 -1 0Source: SEB Enskilda, Datastream and Thomson
Nordea (Buy) has the smallest exposure to Swedish tenant-owner’s associations and their tenant-owners of the Swedish banks with around 40% upside to our fair value estimate
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
23 13 June 2011
The valuation of the Nordic banks peer group is outlined in the following table.
Sector Comment Weakest link in Swedish housing market
28 13 June 2011
Appendix I The following table is based on Boreda’s unique database on the financials of Swedish tenant-owner’s associations. It is based on the most recently released P&Ls and balance sheets and captures 11,400 tenant-owner’s associations responsible for roughly 85% of borrowing by such entities in Sweden. Each bracket includes 570 tenant-owner’s associations.
Detailed data on Swedish tenant-owner's associations sorted by debt per sqm
Sector Comment Weakest link in Swedish housing market
29 13 June 2011
Target prices and risks Target price definition and associated risks Our target price is the analyst's assessment of what total return an investor should expect over the coming six to 12 months. The target is based on fundamental equity research and other factors at the analyst's discretion. Please refer to published reports on the individual companies for a detailed description of the target price methodology.
Risk levels The risk level is the analyst’s view of the uncertainty in the earnings forecasts based on an assessment of the company’s business model, operating risk as well as financial risk. We use two risk levels with the following explanations:
Normal risk: All forecasts involve uncertainty and we view companies in this risk level to have normal forecast risks
High risk: The earnings forecasts are more uncertain than for an average listed company due to business model, operating risk, financial risk or any other reason at the analyst’s discretion. All companies with shorter track record than 12 months as a listed company are by definition classified as high risk according to SEB Enskilda.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
30 13 June 2011
Authors' statement of independence (Analyst Certification) We, the authors of this report, hereby confirm that notwithstanding the existence of any such potential conflicts of interest referred to herein, the views expressed in this report accurately reflect our personal views about the companies and securities covered. We further confirm that we have not been, nor are or will be, receiving direct or indirect compensation in exchange for expressing any of the views or the specific recommendation contained in the report. The preparer of this research report is not registered or qualified as a research analyst, representative or associated person under the rules of the FINRA, the New York Stock Exchange any other US regulatory organization or the laws, rules or regulations of any State in the USA.
This statement affects your rights This report is confidential and may not be reproduced, redistributed or republished by any recipient for any purpose or to any person.
Recipients SEB Enskilda Research is prepared, approved, and issued by Skandinaviska Enskilda Banken AB (publ) (“SEB”), a bank organized under the laws of the Kingdom of Sweden, on behalf of itself and its affiliates for institutional investors. SEB is not a registered Broker-Dealer under the US Securities and Exchange Act of 1934. This report must not be distributed to retail clients except for to retail clients of SEB Enskilda AS ('ESO') located in Norway and, in case of research distributed via Gyllenberg Private Bank in Finland, to its retail clients located in Finland.
Use This research report is produced for the private information of recipients and SEB is not soliciting any action based upon it. If you are not a client of ours, you are not entitled to this research report, and should destroy it. The document is not, and should not be construed as, an offer to sell or solicitation of an offer to buy any securities. Opinions contained in the report represent the author's present opinion only and may be subject to change. In the event that the author's opinion should change or a new analyst with a different opinion becomes responsible for our coverage of the company, we shall endeavour (but do not undertake) to disseminate any such change, within the constraints of any regulations, applicable laws, internal procedures within SEB, or other circumstances. If you are in doubt as to the meaning of the recommendation system used by SEB in its research, please refer to the "Disclaimer" section of SEB Enskilda Research Online, to which our clients are granted access.
Good faith and limitations All information, including statements of fact, contained in this research report have been obtained and compiled in good faith from sources believed to be reliable. However, no representation or warranty, express or implied, is made by SEB with respect to the completeness or accuracy of its contents, and it is not to be relied upon as authoritative and should not be taken in substitution for the exercise of reasoned, independent judgement by you. Recipients are urged to base their investment decisions upon such investigations as they deem necessary. To the extent permitted by applicable law, no liability whatsoever is accepted by SEB for any direct or consequential loss arising from the use of this document or its contents.
Distribution This research report has been prepared by SEB or its affiliates and is being distributed by SEB Enskilda offices in Stockholm, Copenhagen, Oslo, London, Helsinki and Frankfurt. It is being distributed in the United States by SEB Enskilda Inc. ('SEBEI'), a wholly owned subsidiary of SEB organized under the laws of the State of Delaware, USA. Any United States institutional investor receiving the report, who wishes to obtain further information or to effect a transaction in any security discussed in the report, should do so only through SEBEI and not its affiliates. Addresses and Phone numbers for each office can be found at the end of the report.
The SEB Enskilda Group: members, memberships and regulators SEB Enskilda is a trading name used by SEB Merchant Banking Division, SEBEI, and ESO at its offices in Copenhagen, Frankfurt, Helsinki, London, New York, Oslo and Stockholm. Through these SEB Enskilda offices, SEB is a member of inter alia Nasdaq OMX Nordic, the London Stock Exchange, NYSE Euronext , SIX Swiss Exchange, Frankfurt Stock Exchange as well as certain European MTF’s such as Chi-X, Turquoise, BATS Europe and Burgundy. SEB Enskilda uses both the trading identity ENS and SEB. SEB is regulated by Finansinspektionen in Sweden and, for the conduct of investment services business, in (i) Denmark by Finanstilsynet, (ii) Finland by Finansinvalvonta, (iii) Germany by Bundesanstalt für Finanzdienstleistungsaufsicht and (iv) the UK by the Financial Services Authority. ESO is a member of the Oslo Stock Exchange and is regulated by Finanstilsynet. SEBEI is a U.S. broker-dealer, registered with the Financial Industry Regulatory Authority (FINRA). SEBEI and ESO are direct subsidiaries of SEB. SEBEI is not authorized to engage in the underwriting of securities and does not make markets or otherwise engage in any proprietary trading in any securities.
Prevention and avoidance of conflicts of interest All research reports are produced by SEB Enskilda’s Research department, which is separated from the rest of its activities by an Information Barrier; as such, research reports are independent and based solely on publicly available information. Following standard practice, recommendations and target share prices are removed from research on companies which are the subject of public offers on which SEB is advising. The remuneration of staff within the Research department is determined exclusively by research management and senior management (not including investment banking) and may include discretionary awards based on the firm’s total earnings, including investment banking income; however, no such staff receive remuneration based upon specific investment banking transactions. SEB Enskilda’s Compliance department monitors the production of research and the observance of the group's procedures designed to prevent any potential conflicts of interest from affecting the content of research; the latter are described in greater detail in the "Statement of Policies for dealing with potential conflicts of interest surrounding our Research activities" which is available on our Research Online website.
SEB ENSKILDA
Sector Comment Weakest link in Swedish housing market
31 13 June 2011
Your attention is also drawn to the fact that: The current market price of the securities shown in this report is the price prevailing at the close of the business day preceding the date of publication, save where such price was more than 5% different from the price prevailing as at the time of publication, in which case it is the latter.
Unless explicitly stated otherwise in this report, SEB Enskilda expects (but does not undertake) to issue updates to this report following the publication of new figures or forecasts by the company covered, or upon the occurrence of other events which could potentially have a material effect on it.
Specific disclosures for US institutional investors The securities discussed in this research report may not be eligible for sale in all States or countries, and such securities may not be suitable for all types of investors. Offers and sales of securities discussed in this research report, and the distribution of this report, may be made only in States and countries where such securities are exempt from registration or qualification or have been so registered or qualified for offer and sale, and in accordance with applicable broker-dealer and agent/salesman registration or licensing requirements.
The analysis contained in this research report is based on numerous assumptions; different assumptions could result in materially different results. Any valuations, projections and forecasts contained in this report are based on a number of assumptions and estimates and are subject to contingencies and uncertainties. The inclusion of any such valuations, projections and forecasts in this report should not be regarded as a representation or warranty by or on behalf of the Group or any person or entity within the Group that such valuations, projections and forecasts or their underlying assumptions and estimates will be met or realized. Past performance is not a reliable indicator of future performance. Foreign currency rates of exchange may adversely affect the value, price or income of any security or related investment mentioned in this report. In addition, investors in securities, such as ADRs, whose values are influenced by the currency of the underlying security, effectively assume currency risk.
Company specific disclosures and potential conflicts of interest SEB Enskilda acted as Joint Lead Manager and Bookrunner of the secondary placement of shares in Nordea (on behalf of Kingdom of Sweden) in February 2011. This report has been produced by SEB Enskilda's Research Department, which is separated from its Corporate Finance Department by an information barrier; as such, it is independent and based solely on publicly available information. A member of, or an entity associated with, SEB or its affiliates, officers, directors, employees or shareholders of such members (a) has never been represented on the board of or similar supervisory entity of the companies mentioned in the report, (b) has from time to time bought or sold the securities issued by the companies referred to in this report or options relating to these companies, and (c) SEB or its affiliates beneficially own less than 1 % of a class of common equity securities of the remaining companies mentioned in the report, as of 10 Jun 2011. Within the past 12 months SEB Enskilda has been involved in a public offering and/or underwriting of Nordea. The analysts responsible for this report (jointly with their closely related persons) may hold shares or other instruments related to the companies mentioned in this report. Please refer to published reports on the individual companies for details. The information can also be found on our website, at the following address: www.sebenskilda.se.
SEB Enskilda’s standardised recommendation structure Consolidated distribution Corporate Finance clients last 12M. as per 31 Mar 2011 (%) Distribution as per 31 Mar 2011 (%) Buy Attractive risk/reward - at least 10% upside to target price. 61.5 11.2 Hold Fairly valued - the shares are trading close to target price. 34.5 3.7 Sell Unattractive risk/reward - the shares are trading above target price. 4.0 0.3 Unrated Company not covered, or we are not allowed to have a recommendation for compliance reasons.
Skandinaviska Enskilda Banken AB (publ). All rights reserved.
Copenhagen Bernstorffsgade 50 P.O. Box 2098 DK-1577 Copenhagen V