Seeking Alpha
What is your profession? ×
Profile| Send Message|
( followers)

The following is a follow up article about the Australian housing and banking sectors. This one focuses specifically on the individual bank fundamentals. It is a bit long for this format so if you would rather read it in Scribd format feel free to do so here.

In their fifth annual international housing affordability survey based on Q3 2008 data, researchers Wendell Cox of Demographia.com and Hugh Pavletich of Performance Urban Planning ranked 265 separate markets in Australia, New Zealand, Canada, Ireland, the UK and the US in terms of affordability.

As has become painfully obvious, some of the largest housing bubbles have proliferated in the English speaking world. However, the decline in prices has been much more acute in the US than in Australia (AU) and New Zealand (NZ), as evidenced by the index data I presented in my previous piece on this subject. The risk of this ongoing discrepancy is as follows: if housing prices in AU and NZ are unsustainable then the banks who lent money people to buy these overvalued houses could be in for some serious losses, similar to what has happened in the US.

So, what did Mr. Cox and Mr. Pavletich find in their survey? Using a scale in which 3.0 or less is affordable, 4.1-5.0 is seriously unaffordable, and 5.1 and over is severely unaffordable, the researchers determined that 24 of the 64 severely unaffordable markets were in Australia. Additionally, 3 out of the 40 seriously unaffordable markets were in Australia and no markets were deemed moderately unaffordable or affordable.

Going through the list of the top 20 least affordable markets in the survey, 8 are located in AU with the #1 spot belonging to the Sunshine Coast of Queensland (an amazing 9.8 ranking) and the #3 spot to the Gold Coast of Queensland (8.7 ranking). In fact, the median score for the Australian cities was a 6.0, versus 3.5 for Canada, 5.4 for Ireland, 5.7 for the UK and 3.2 for the US. Commenting on the AU housing markets the researchers conclude:

Unlike the other national markets in the Survey¸ Australia has thus far been able to avoid material house price declines. It seems likely that, sooner or later, the inherent instability and unsustainability that characterizes bubbles will lead to house price declines in Australia. However, were it possible for Australia to retain its highly over-valued house prices, there would still be a significant cost. Future generations would pay far more for housing than in the past, and Australia’s relative standard of living would decline.

Furthermore, the results for NZ are not a whole lot better. Out of the 8 total markets surveyed, 7 were deemed severely unaffordable and 1 was noted as seriously unaffordable. As a result, the median ranking for NZ was 5.7, just slightly below that of AU, although no markets were able to crack the dubious top 20 least affordable markets.

In comparison, the cities considered the most expensive in the mainland US almost look like bargains in relation to some markets in the AU-NZ regions. The most expensive market in the US was Honolulu (9.1), followed by San Francisco (8.0), San Jose, CA (7.4), San Luis Obispo, CA (7.3) and Los Angeles (7.2). Even notoriously expensive places to live such as New York (7.0) and London (6.9) look cheap compared to AU cities like Sydney (8.3).

Bank Valuation and Fundamentals

Now, Cox and Pavletich do go on to discuss the reasons why these markets have become so unaffordable and the piece is actually very interesting for people who want to understand how housing bubbles develop.

However, I want to turn my focus onto the banks themselves. So far in my two analyses I think I have done a reasonably good job of illustrating the magnitude of housing price increases in the AU-NZ region and discussing why some of these prices may not be sustainable.

But the most important questions to try to answer are what will be the ultimate effect on the banks of a fall in real estate prices and do their current valuations reflect the potential damage to their balance sheets?

Data from company filings and my calculations

I present the above chart (based on the most recent data available) to illustrate how exposed these companies are to the seemingly irrationally exuberant housing markets in AU and NZ. What this data tells me is that these banks will perform only as well or as poorly as the credit tied to housing loans.

Therefore, the first thing I looked was the credit metrics of the banks. In tough times it is especially important to see that banks have put aside the necessary reserves for potential loan losses and to evaluate the degree to which credit quality has been deteriorating. At least up until the most recent data released by these seven banks, they appear to score well on both of those metrics.

Data from company filings and my calculations

It is obvious that the global recession and initial slide in housing prices have had some negative effects on credit. Over the previous five years the average ratio of non-performing loans to total loans was only 28 basis points (bps). As a result of the recent turmoil that average has jumped to 80 bps, but is significantly below that of even some of the more conservative and credit worthy US banks.

The only one that stands out is SUN with its 1.77% NPL ratio. This is up from .64% in June 2008 and is well above the average from the previous 5 years. This larger deterioration in credit may be why shares of SUN trade at a larger discount to historical multiples (see valuation chart below) than its competitors. On the other hand, Bank of Queensland (BOQ) looks to be the most conservative underwriter of the group as its current and historical NPLs ratios are the lowest.

This may also account for why BOQ has the lowest average ROA and second lowest average ROE over the last five years (see return chart below). Additionally, I am always dubious of a bank that does not have large enough reserves to cover net charge offs and BOQ fits that bill.

However, by in large these banks are very conservatively reserved. These days it is rare to find a US bank with $1 of reserves for each $1 of NPL, a sign that the bank is proactively preparing for future losses. However, all the banks other than Suncorp-Metway (SUN) (which has very low net charge offs and has a large insurance subsidiary that may influence the need for large reserves) are at or near 100% in terms of Allowance/NPLs.

These robust allowances allow the banks to weather a storm of new NPLs without suffering outsized effects on net income or the balance sheet (as provisions reduce equity and tangible equity). When the June data comes out for the banks with fiscal years that end in June we will see the extent to which credit has deteriorated further.

But, as of right now, aside from the relatively low allowance ratio for SUN and the comparatively high net charge off ratio for Australia & New Zealand Banking Corp. (ANZ) nothing about the credit metrics really stands out as problematic.

6/29/2009

Price/Earnings

Price/TBV

Price/BV

Current

5 Yr Average

Current

5 Yr Average

Current

5 Yr Average

Australia & New Zealand Banking Group

12.66x

13.24x

1.44x

2.78x

1.28x

2.33x

Bank of Queensland

13.89x

16.92x

1.64x

2.34x

1.00x

2.11x

Commonwealth Bank of Australia

11.28x

14.98x

2.64x

3.61x

1.93x

2.45x

Suncorp-Metway Ltd

14.27x

13.40x

1.22x

3.03x

0.53x

2.06x

National Bank of Australia

9.24x

13.40x

1.52x

3.03x

1.23x

1.96x

Bendigo-Adelaide Bank

13.82x

17.37x

1.59x

2.35x

0.72x

2.21x

Westpac Banking Corp.

11.92x

14.29x

2.44x

3.21x

1.67x

2.68x

Group Average

12.44x

14.80x

1.78x

2.91x

1.19x

2.27x

Data from Cap IQ and my calculations

This second chart compares the current valuations of the banks to their 5 year averages and to the group averages. Just about across the board and on every metric these banks are all trading at multiples below their 5 year averages. Of course, based on what has happened in the global economy and stock markets, this is not a surprise.

Of note, while the banks are still trading well above tangible book value, the multiples have contracted significantly as compared to the 5 year averages. The question going forward is whether or not these banks will have the loan book growth, EPS growth and returns on equity to justify trading at close to a 3x TBV multiple (2.91x is the 5 year average for the group).

In truth, solely based in relative and comparative multiples, none of these valuations really stand out as being extremely over or undervalued. Based on these valuations two of the harder hit banks on a TBV basis are SUN and Bendigo-Adelaide (BEN) while it appears that Commonwealth Bank of Australia (CBA) has been spared to some extent. This can also been seen in the chart below:

% Above 52 Week Low

% Below 52 Week High

Australia & New Zealand Banking Group

37.02%

21.50%

Bank of Queensland

41.11%

45.73%

Commonwealth Bank of Australia

61.96%

22.19%

Suncorp-Metway Ltd

47.25%

56.42%

National Bank of Australia

41.14%

27.35%

Bendigo-Adelaide Bank

16.72%

51.61%

Westpac Banking Corp.

37.36%

20.31%

Group Average

40.37%

35.01%

Data from Google Finance and my calculations

While SUN has had a nice bounce off of its 52 week low, it still has suffered the largest fall from the high. BEN, on the other hand, experienced the most dramatic decline and has barely bounced off the bottom. In contrast to the preceding two banks, shares of CBA have rallied substantially off their lows and are now only off their 52 week high by about 22%. Still, nothing about this chart is particularly outstanding as most of the data is centered close to the mean.

Accordingly, when valuations by themselves are not especially intriguing on the long or short side the next step is to look at the returns of these banks. On the long side investors should look for banks that have historically strong ROEs but are trading at reasonable multiples to BV and TBV. Conversely, on the short side, bank bears should focus on companies with meager ROEs that are trading at growth-like multiples.

TTM ROE

5 Year Avg. ROE

TTM ROA

5 Year Avg. ROA

1H 2009 NIM

Australia & New Zealand Banking Group

10.40%

17.06%

0.60%

1.08%

2.22%

Bank of Queensland

6.30%

12.58%

0.40%

0.68%

1.52%

Commonwealth Bank of Australia

18.30%

17.18%

0.90%

1.04%

2.04%

Suncorp-Metway Ltd

3.30%

14.70%

0.50%

1.42%

1.84%

National Bank of Australia

14.30%

15.98%

0.70%

0.88%

2.07%

Bendigo-Adelaide Bank

4.70%

12.30%

0.30%

0.74%

1.49%

Westpac Banking Corp.

14.80%

19.30%

0.80%

1.04%

2.35%

Group Average

10.30%

15.59%

0.60%

0.98%

1.93%

Data from Cap IQ and my calculations

The first thing to notice about the previous chart is the meager return on asset ratios for these banks as compared to relatively robust return on equity ratios. What this indicates (and will be shown in much more detail later) is that these banks are highly levered. The denominator of the ROE calculation is so small that these companies generate nice returns on equity but their asset base is so large that ROAs are much less attractive than those of their US-based brethren.

What this means is that ROE can be somewhat deceiving. Specifically, the amount of leverage the banks with the highest ROEs could potentially add incremental risk to an investment. In any case, from looking at the historical returns, it would appear that Westpac Banking Corp. (WBC) and CBA are the best in class when it comes to ROE and CBA, WBC and ANZ are the best of the bunch when it comes to ROA. It should be mentioned that since SUN has a large insurance subsidiary that contributes a large portion of its revenue, its ROA may not be perfectly comparable to that of its peers.

Finally, when it comes to net interest margin (NIM), I have to assume that the reason the banks have such low NIMs when compared to their US counterparts has to do with either more expensive funding costs or lower average rates on their loans. In either case the industry as a whole has consistently had a lower NIM than most US banks. However, the banks that seem to be the best situated on this metric are WBC and ANZ.

Now the question is to what extent these returns are based on excessive leverage?

Gross Loans/TE

Gross Loans/Deposits

TE/TA

Australia & New Zealand Banking Group

14.31x

1.18x

4.87%

Bank of Queensland

29.07x

1.20x

2.91%

Commonwealth Bank of Australia

20.98x

1.29x

3.53%

Suncorp-Metway Ltd

10.49x

2.38x

5.89%

National Bank of Australia

14.09x

1.23x

4.12%

Bendigo-Adelaide Bank

30.29x

1.17x

2.71%

Westpac Banking Corp.

19.24x

1.41x

4.04%

Group Average

19.78x

1.41x

4.01%

Data from company filings and my calculations

I think this is probably the most interesting difference between the AU-NZ banks and the US banks. I don’t think I have ever come across a US regional bank with a gross loan to tangible equity (TE) ratio above 20x.

Just for a quick reference point, as of Q1 2009, US-based Zions Bancorp (NASDAQ: ZION) had a gross loans/TE multiple of 14.93x, Western Alliance Bancorp (NYSE: WAL) a 16.57x multiple and City National (NYSE: CYN) a conservative 10.73x multiple. All three of these banks had ratios much lower than the group average for the AU-NZ banks. The discrepancy becomes even worse if SUN is taken out of the mix (due to its formidable insurance operations). In that case the average for the AU-NZ group jumps to 21.33x.

When I evaluate US banks, I am looking for a minimum tangible equity to tangible asset ratio (TE/TA) of 5%. This cushion gives me confidence that a bank has at least a chance (depending on its loan book of course) to make it through this downturn without raising an exorbitant amount of dilutive capital.

Again, taking SUN out of the equation, only ANZ comes even close to meeting that threshold. In fact, when it comes to TE/TA BEN and BOQ look undercapitalized. They both have TE/TA ratios under 3% and have a gross loans to TE ratios over an unheard of in the US (unless you include Bear Stearns and Lehman as banks) 29x. In short, investors should be careful when scrutinizing the ROEs of these companies. Ideally, the best banks to buy have strong ROEs but relatively less leverage than their peers. On the other hand, highly levered banks with below average ROEs such as BOQ could be interesting on the short side.

The AU-NZ Housing Markets

The levered nature of the banks’ balance sheets leads to the most important point about the AU-NZ housing markets. From my perspective, the only reason it could make sense for these banks to be so highly levered would be if the losses they suffer when housing loans go bad are much less than those that US banks suffer.

Since I finished the preliminary article on the banks, I have decided to do some more digging. One of my readers (many thanks to Sujie Huo) steered me to the site of an AU hedge fund manager who lives in Sidney and has intricate knowledge regarding the Australian housing market and economy. He is a former banking and insurance analyst who wrote these very prescient words in June of 2007:

Mortgage debt outstanding in the US has grown by almost 600 dollars per household per month…Simple math tells you that it cannot go on forever. Herb Stein was an economics advisor to President Nixon, but he is best remembered for asserting that “things can’t go on forever”…For several years we have been searching for a trigger that might force Stein’s law to apply with more urgency but have yet to find one. The subprime mortgage mess is likely a trigger as any…

The insight I have been able to gain through reading his blog, exchanging emails and speaking with him is not easily available or verifiable using primarily Google searches. Specifically, the most useful information I have gleaned from this fund manager is that in Australia many housing loans are recourse.

This means that buyers are personally liable for the paying back the loan. By in large this is not true in the US for housing loans. On many commercial real estate deals banks do try to get personal guarantees but they are very rare in home mortgage agreements. What this means in the US is that the only recourse the bank has if a person does not live up to his or her end of the bargain is to foreclose on the property.

At that point the bank owns the asset (house, apartment, condo) and then can resell it. However, after including the legal costs of foreclosure, any repairs needed, and brokers/auction fees, banks are often forced to realize severe losses when they foreclosure. This is in addition to the sales price that is often below the value of the mortgage, based on the recent dramatic decline in prices across the US.

In contrast, in Australia, banks are often sheltered from losses due to the fact that the recourse nature of the loans incentivizes people to sell prior to an actual foreclosure. Interestingly enough, unlike in the US where auctions are used for distressed properties, in AU auctions are the primary method employed to sell a house and the only outright sales are tied to distressed borrowers. In addition, since the borrowers are on the hook for the entire amount of the loan and therefore care about the eventual sale price, they traditionally make sure that the property is kept in good condition.

Unfortunately, this is not the case in the US. It is not hard to find articles in the popular press about the neglect and abuse houses that are soon to be foreclosed upon suffer as the defaulting borrowers basically squat in the house until the sheriff knocks on their door.

In addition, I got some more information about loan to value policies in Australia the fund manager. According to him, none of the banks offer mortgages with an LTV over 80% without requiring personal mortgage insurance (PMI). Apparently there were some 100% LTV mortgages originated that were ultimately securitized, but since that market has dried up there is very little activity among these higher risk loans.

Accordingly, while a housing bust might cause many existing mortgages to be underwater, the recourse nature of the loans as well as the PMI on the higher LTV agreements will limit the banks losses (as long as the PMI counterparties are still solvent enough to live up their obligations). Furthermore, the manager believes that the AU government would not let any of the banks go bust and the market understands this implicit guarantee.

Therefore, unlike in the case of Bear Stearns, it is unlikely that a short term funding shortfall would be the death of any of the banks. He even suggested that the AU government is so solvent it could potentially guarantee all the debt issued by the banks without destroying the country’s balance sheet.

In general, while he knows housing prices in AU are basically insane, he thinks the banks margins are so strong and government backing so credible, that even a 50% decline in prices would not be fatal to any of them. When I asked him about the weaknesses of the banks he listed a few. First, the banks are not particularly closely tied to the strong mining industry. Since most of the loan portfolios are tied to housing, the banks only indirectly benefit from a robust export-based economy.

Also, he admitted that there was some currency risk for US-based investors who are keen on betting on a continued appreciation of the AU dollar versus to US dollar. He indicated that if China begins to struggle the AU dollar could fall significantly in the short run. But he believes that China would then buy as many AU dollars as it could as a claim on the value of AU’s vast natural resources. Therefore, in the long run China would support the AU dollar.

Finally, when I asked the manager to identify a potential catalyst for the housing bust that he feels will eventually come he suggested that the NZ loan books could be the first domino. He envisions credit deterioration in the NZ market causing the AU banks to pull in their horns in AU, a move that would create a credit contraction that reduces the supply of money for home loans While the appreciation in AU housing prices has been much more drawn out affair than the US boom that took off in the early part of this decade, it does not necessarily mean that a rapid reversal is not possible.

The reason he sees NZ as the problem is that the country is nowhere near as resource rich and unless agricultural commodities really took off there would be no way for the NZ economy to rebound. Thus, the eventual L-shaped “recovery” and housing price spiral (remember NZ has seen a larger increase in house prices than the US and AU since 1999) could be what ultimately causes house prices in AU to fall as well.

What Does It All Mean?

The fund manager’s conclusion is that these banks are very dubious shorts for the reasons outlined above. He just doesn’t believe that betting against the banks is the best way to benefit from the inevitable real estate downturn that will eventually hit AU and NZ.

Based on my analysis, I tend to agree with him with one exception. None of the stocks stands out as a screaming short based on valuation, returns or credit trends but there is one that looks the weakest. Specifically, when I evaluate banks as potential shorts I try to identify valuations that are not justified by the fundamentals. The only bank that seems to combine a high valuation with low levels of capital and reserves, as well as low returns, is BOQ. BOQ has the second lowest TE/TA ratio, the second highest gross loans to TE ratio, the second lowest allowance to NPL ratio, the second lowest average ROA and ROE, the second lowest 1H 2009 NIM, and is the only bank whose allowance does not cover charge offs.

In addition, BOQ had close to 72% of its loan book tied to housing as of the latest available data, well above the 60% average for the group. Also, it is important to remember that 65.4% of BOQ’s loans as of August 2008 were located in Queensland, the region that had 2 of the top 3 most expensive housing markets as ranked by researchers Cox and Pavletich.

Despite this, BOQ is trading at the second highest P/E ratio (based on trailing 12 month earnings) and the third highest price to tangible book ratio. But, as mentioned above, BOQ historically has had the lowest percentage of NPLs and as of the most recent filing had the lowest current NPL ratio.

Accordingly, it is a difficult short call because it is not easy to determine whether the relatively higher P/E ratio and the lower capital levels are appropriate given the bank’s underwriting history. However, some news out Friday calls into question the legitimacy of BOQ’s more robust multiples. This article in The West Australian indicates that BOQ could be heavily exposed to losses stemming from involvement with a failed financial group named Storm Financial.

BOQ was caught off guard last week when the Australian Securities and Investments Commission (ASIC) placed the bank under investigation for matters relating to Storm one day after BOQ issued a denial over a regulatory probe.

Plaintiff law firm Slater and Gordon is preparing litigation against the bank.

One of the law firm's principals, Damian Scattini, said there would be a series of cases against BOQ that left the bank with "huge" financial exposure.

"It's a very large exposure that they have and they know it and we know it, but they're just not publicly saying it," he said.

Asked if the exposure could reach into the hundreds of millions of dollars, Mr Scattini said it was a very significant problem for the bank.

The article seems to indicate that a settlement is possible but I see this as incremental headwind that might put additional pressure on the shares of BOQ.

Unfortunately, there are also some structural issues with shorting these shares. First, since they all trade on the Australian Stock Exchange, there could be high borrowing costs and commissions that reduce the return. Also, the smaller banks are not particularly liquid, a circumstance that could potentially lead to a painful short squeeze. Furthermore, there is a good deal of short term risk due to the reflation trade lifting prices of commodities and thus pushing up shares of all companies with exposure to the AU economy.

Finally, adding to the risks and costs of shorting the AU banks are the following dividend yields that must be paid by the short seller. As the chart below illustrates, these banks continue to pay non-trivial dividends that obviously could be cut at any time (in order to preserve capital) but in the meantime could limit the return on a short.

Annualized

Dividend Yield

Australia & New Zealand Banking Group

5.68%

Bank of Queensland

5.85%

Commonwealth Bank of Australia

5.81%

Suncorp-Metway Ltd

6.23%

National Bank of Australia

6.53%

Bendigo-Adelaide Bank

8.27%

Westpac Banking Corp.

5.66%

Group Average

6.29%

Data from company filings and my calculations

As a result of the difficulty in shorting these banks, the question arises about whether or not any of these banks could be interesting on the long side. From what I can tell after comparing returns, capital levels, allowances, and valuation, the best in class of the group appear to be CBA and ANZ.

While NAB also has some attractive metrics, the bank has significant exposure to off balance sheet special purpose entities (SPEs) that makes me very hesitant to recommend a more in depth analysis of the stock. CBA on the other hand has so far been able to avoid severe credit deterioration and as of the December 2008 filing had NPLs of only 60bps versus a 5 year average of 17bps.

Additionally, CBA had the second highest allowance to NPL ratio and its allowance covered charge offs by more than 7 times. Over the last 5 years CBA has had the second highest ROE of the group and one of the higher ROAs.

Furthermore, CBA also has a large wealth management business in which the company manages over $100 billion AU dollars of client assets and subsequently has not been nearly as dependent on interest income as its rivals. Specifically, on average, interest income as a percentage of revenue over the past 5 years has been 49.25% versus 64.2% for the group (excluding SUN). Plus, CBA has minimal exposure to the at-risk NZ housing market (close to 86% of CBA’s loans are tied to AU).The problem is that CBA trades at the highest multiples to book and tangible value among the banks and trades at a smaller discount to its 5 year average multiples than its peers.

When this data is combined with the fact that CBA has the third highest gross loan to TE ratio and the third lowest TE/TA ratio, the stock appears to be fairly valued, especially given the potential headwinds facing the housing market.

The other company that could be a compelling long, ANZ, has historically had the second highest ROE of the group and in the most recent filing period had the second highest NIM. While ANZ had the second highest NPL ratio according to the latest data, the bank’s allowance to loan loss ratio was over 110% and even elevated net charge offs of 43bps were covered 2.76 times by the provision. ANZ also had the second lowest gross loans/TE ratio and second highest TE/TA ratio among the seven banks.

Despite sufficient reserves and low levels of leverage ANZ trades at less than 55% of its 5 year average price to book and price to tangible book ratios. Also, ANZ only has 53% of its loan book tied to housing but unfortunately has over 23% of its portfolio exposed to NZ. In trying to balance the positives and negatives regarding ANZ (and keeping in mind the fact that there could be a large housing bust on the horizon), it looks like ANZ is fairly valued at this level as well. At very least, similarly to shares of CBA there is not a margin of safety built into the current price.

Potential Investment Strategies

Having identified the best in class banks and established that they are not overwhelmingly compelling at the current levels and after highlighting the difficulty in shorting any of the banks, should I conclude that investors are well served to forget about the AU banks? Absolutely not. I think there are a number of strategies that could turn out to be profitable for patient investors:

  1. Wait to buy ANZ and CBA on a pull back: Assuming that the balance sheets of these two banks do not deteriorate significantly as a result of a severe decline in housing prices, both become much more interesting if they decline in line with the general stock market or as a result of macroeconomic concerns. Obviously the housing market and economic conditions in the AU-NZ region need to be watched carefully, but investors should have a valuation in mind that would prompt careful consideration of a purchase. CBA hit a 52 week low of $24.03 in January of 2009, at which point the stock was trading at about 1.63x December 2008 tangible book value. For a company with a solid historical ROE such a depressed valuation seemed unwarranted. Accordingly, if CBA fell back down to around 2x tangible book value or $28.50 a share, the company would be worth another very close look. At that point the dividend yield would be over 9% and even if it were cut in half it would still be attractive. In terms of ANZ, the stock hit a 52 week low of $11.83 in February of 2009. At that price the stock would have been trading at 1.05x March 2009 tangible book value. If the stock traded around tangible book or around $12 again, the dividend yield would be over 7.5% and would likely provide satisfactory risk adjusted returns for an investor willing to buy and hold.

When it comes to liquidity, according to Google Finance ANZ trades over 7.5M shares a day CBA over 4.1M so liquidity should not be an issue. Also, I think it should be mentioned that as of the most recent filings, ANZ and CBA had dividend payout ratios of 71.3% and 65.3%, respectively, a fact that indicates that both companies have been comfortably covering their dividends.

Also, while it is tempting to consider putting together a basket trade in which a number of these bank stocks are purchased, I do not see this as a viable option. While it does limit company specific risk and exposure, my concern is that the returns on the weaker banks will either dilute or offset the returns from the stronger banks. Just like what has occurred in the US with the stronger banks such as JP Morgan (NYSE: JPM), I would not be surprised if the returns on the better AU banks decoupled from those of the weaker banks.

As a result the only way I think it makes sense to diversify would be to buy ANZ and CBA and position size the investment as if it were one position.

  1. Execute a pair trade with a long position on CBA and/or ANZ and a short position in BOQ: By going long one of the best in class bank and short the weakest of the group an investor could be protected from stock price declines occurring as a result of weakness in the AU-NZ housing market. This pair trade makes sense because BOQ’s loan book is much more exposed to housing and is tied to an apparently very expensive Queensland market. The risk of course is that BOQ is already down close to 46% from its 52 week high and either falls less than ANZ or CBA in a market downturn or goes up more in an upturn. However, there is also the potential for investors to be rewarded based on a further delineation between the strongest and weakest banks. In this scenario the numerous headwinds facing BOQ could send shares down even while shares of the stronger ANZ or CBA appreciate or tread water as investors fly to quality.

The stocks all have current dividend yields that are in the 5.5%-6.0% range so at least the dividends would be offsetting in a pair trade. I would also argue that with a current 2.91% TE/TA ratio and a 106% dividend payout ratio in the most recent period, BOQ is more likely to cut the dividend than CBA or ANZ.

Finally, in terms of liquidity, according to Google Finance BOQ is relatively liquid, trading over 320K shares a day. However, investors should keep in mind that with the stock under $9 that trading volume implies less than 2.9M AU dollars each day. As a result, it is possible that BOQ is too illiquid for some investors or a short position could have to be accumulated over time. (Note: I have not investigated the cost of borrowing BOQ.)

  1. Execute a pair trade with a long position on CBA and/or ANZ and a short position in one of the PMI insurers in AU: The two companies with the largest PMI exposure in AU are QBE Insurance Group (QBE) who bought the AU business from the embattled PMI Group (NYSE: PMI) in 2008 and Genworth Financial (NYSE: GNW). If it is true that that the mortgage insurers will be on the hook for losses as more high LTV loans go bad, shorting either one of these companies could provide an effective hedge against declines in ANZ or CBA. With shares of GNW already having been decimated and trading at .37x book value, QBE trading at over 3.6x tangible book looks like a much better short candidate. In fact, although I haven’t had a chance to dig into the specifics, shorting QBE may be the best way to play the coming housing bust in the AU-NZ region. In comparison to BOQ, QBE is very liquid as it trades over 2.3M shares a day. Also, its 6.3% dividend yield is on par with those of ANZ and CBA so there is an offset available.

Conclusion

Based on all the research I have done I believe house prices in AU and NZ have appreciated to the extent that the entire markets are overvalued. With all of the turmoil in the global economy and the housing busts that have occurred across the English speaking world, it is hard to see how the current prices are sustainable. Accordingly, I would not be surprised to see housing prices fall even further than they already have from their peaks.

Of course, the tight supply in AU and the recourse nature of the loans may limit the decline. However, any fall in housing prices combined with increased unemployment in the region will inevitably lead to more NPLs and higher charge offs.

Thus, I expect the shares of the seven banks to be under pressure in the near term. The best ways to play this potential weakness are outlined above in the Potential Investment Strategy section. While none of these strategies is fool proof, I think they all provide investors with either a margin of safety or hedge against permanent capital impairment.

Disclosure: No positions