13 March 2011

Covered Bonds and Big Four Swedish Banks

When exploring the balance sheets of large European banks, it stands out that Nordic banks have remarkably high Loan-to-Deposit ratios. Take for example Swedish banks (the following data is taken from 2010 annual reports of the mentioned banks). At the end of 2010, Svenska Handelsbanken had a Loan-to-Deposit ratio of 2.71 (i.e. only 36.9% of its loans to the public were financed with deposits). The same numbers for Swedbank were 2.22 (45.0%), for Nordea 1.78 (56.1%) and for SEB 1.51 (66.2%). So, in that sense the balance sheets of these banks seem pretty unbalanced. Covered bonds have played a rather important role in this result. Let’s see how and what are the related risks. 

Covered bonds are debt securities backed by cash flows from mortgages or public sector loans. Compared to other corporate bonds, they have one important enhancement: recourse to a pool of assets (cover pool) that secures or "covers" the bond if the originator (usually a financial institution) becomes insolvent. They are intended to be more creditworthy than non-covered bonds, which reduces financial institution’s cost of funding. Naturally, the cover pool is subject to certain criteria like Loan-to-Value ratios of mortgage loans in it. More information about the covered bonds legislation in Sweden specifically can be found from the website of ASCB (Association of Swedish Covered Bond issuers).

For illustrating the matter with a simple numerical example, let’s assume a bank with a simplified balance sheet as depicted in Figure 1.1 below. Let’s further assume that the bank funds all of its assets with deposits and covered bonds only, that it has to keep 10% of its liabilities as cash reserves in central bank and is able to issue covered bonds up to 50% of loans serving as cover pool.

Based on this simplified balance sheet and the assumptions that we just made, the bank can issue covered bonds in the amount of 4,500 units. If depositors are also buyers of all of these covered bonds, it’s a pretty pointless or even undesirable exercise for the bank: there will be no additional cash. What changes is the funding structure (and it changes towards being more expensive, because bond investors while not benefitting from deposit guarantee, require higher returns) with the negligible positive effect of reducing the risk of bank runs. Thus, the idea of covered bonds is rather to sell at least a part of them to outside investors and involve fresh money.

Figure 1.2 depicts the situation where covered bonds in the amount of 50% of loans (as assumed) are issued to outside investors. Fresh money in the amount of 4,050 units has ended up as available cash for lending (4,500 units minus additional cash reserves).

Now these 4,050 units are lent out as many times as the reserve ratio (10% by assumption) and the respective deposit expansion factor allow (for the calculation logic, see also “Money Creation – Basic Principle”). The result is on Figure 1.3 below. Although having had a dramatic impact to the amount of outstanding loans, Loan-to-Deposit ratio has remained under control while being 98% – and it remains below 100% until we assume that deposits do not leave the bank, and covered bonds are all bought by outside investors (because newly granted loans end up as deposits on the other side of the balance sheet).

Looking real numbers, it isn’t like that. People still do something with their money. For example, they invest it into investment funds, incl. pension funds, which in turn buy covered bonds as a safe investment (typically, covered bonds – at least Swedish covered bonds -- have AAA ratings). In this way, bank’s balance sheet goes (seemingly) out of balance; for illustration, see Figure 1.4, where via investment funds one third of the deposits have ended up in covered bonds. It can easily be calculated that now the Loan-to-Deposit ratio is 147%.

Where are the risks of this kind of scheme involving covered bonds? Well, they arise when there is cheap money outside that is searching for a “safe harbour”.

First and foremost, it’s the rapid credit expansion that covered bonds enable. While being perceived as safe investment (high ratings, overcollateralization, low Loan-to-Value ratios etc.), they attract outside investors and make funding cheap for banks. Easy access to loans drives up mortgage prices, which in turn makes Loan-to-Value (LTV) ratios to look even better and covered bonds seem even safer. The one who bought a flat in Stockholm in summer 2000, had a LTV of her/his loan at 85% at that time, and (hypothetically) hasn’t repaid anything, has now (as of December 2010) an LTV ratio of 42.5% -- just because mortgage prices have doubled since the time of home purchase (again simplified while not taking into account the depreciation, but still illustrative). Going forward, there is an even greater incentive for the banks to increase mortgage lending because precisely this lending is the source of funding.

Secondly, it’s the high and complicated exposure that banks have towards local mortgage market: it’s not only about the quality of lending portfolio and credit losses, but also about price and availability of funding. If the trend should reverse and artificially high mortgage prices start to decline (either because customers have become over indebted or because of some external shock), cover pools begin to decline (because of extensive defaults and because of increasing LTV ratios that make part of the underlying loans ineligible) and funding costs to increase (as rating agencies are downgrading covered bond ratings and investors are much less willing to buy them). Further vicious circle continues as lending becomes more restricted in result of limited cash.

Thirdly, cheap money extracted from local mortgages starts flowing into the other countries and causes large imbalances there (as illustrated e.g. by the Baltic experience). Of course, in the event of some kind of internal or external shock, these imbalances materialise in high credit losses.

The fourth aspect to consider in a downward spiral – as we now know only too well – is public finances, and angry tax payers. No matter if it’s the direct capital injection into the banks or preventing credit customers from defaulting via social safety nets, at the end simple employees are those who have to pay for this. Problems in financial sector, declining share prices and bond prices also further aggravate problems related to the demographic situation (among others, people’s investments in pension funds decline).

Considering the above described risks and the proportion of funding that four major Swedish banks do based on Swedish mortgages (see Figure 2 below), it’s no wonder why the governor of Riksbank, Stefan Ingves, seems so worried about risks in Swedish mortgage market and why he stresses the importance of Basel III. Looking at Figure 2, it’s also no wonder why eligibility of covered bonds for liquidity buffer is that important for Swedish banks (in the context of Basel III, and Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR)) – otherwise they would need to significantly revise their current funding strategies and business models.

To conclude, cheap money outside is dangerous and covered bonds tend to attract it. Accepting covered bonds as eligible for liquidity buffers would be a mistake from systemic perspective; it would make covered bonds even more attractive and boost mortgage lending further. The safe harbour of Swedish mortgages is not that safe as it looks, and Swedish major banks are highly and in a complex manner exposed to it.


  1. Banks have a much greater array of financial techniques to allow them to better manage their exposure to their funders and maintain their growth in lending, despite having a relatively high LTD ratio.

  2. Thanks for another informative site. Where else may I am getting that kind of info written in such a perfect means?
    I've a venture that I am simply now working on, and I
    have been at the look out for such info.

    Feel free to visit my web site; teleselskaber

  3. I have been browsing online more than 3 hours lately, yet I never found any attention-grabbing article like yours.
    It's pretty value sufficient for me. In my opinion, if all webmasters and
    bloggers made good content as you probably did, the web will likely be much more helpful than ever before.

    Here is my homepage - mobilabonnement priser

  4. Hi there, this weekend is pleasant in support of me, because this time i
    am reading this wonderful educational article here at my home.

    Feel free to surf to my web site billig mobilabonnement