To print this update, download a PDF version.
After the carnage that characterised global share markets in the first two months of the year, March saw solid recovery as most major markets bounced back strongly. This confidence boost is in part because of news that China’s economy has stabilised in the short term, and senior officials from the US Federal Reserve assuring investors that interest rates will remain lower for longer.
It’s worth noting that at the time of writing, the US major share price index is actually higher than where it started the year, despite being down by more than 10% in mid-February. While Australian shares also enjoyed the March rally, this market remains in negative territory. A key reason for this underperformance is weakness in Australian bank shares, which are now being ‘short sold’ in record volumes. In this month’s update we delve into what’s driving sellers’ behaviour and whether it is warranted.
Performance of key markets
||3 Years p.a.
||5 years p.a.
|Australian Shares (S&P/ASX 300)
|US Shares (S&P 500) in US Dollars
|US Shares (S&P 500) in Australian Dollars
|Asian Shares (MSCI All Country Asia ex Japan)
|Australian Dollar (AUD/USD)
|Australian Fixed Interest (Bloomberg Composite)
|Cash (Bloomberg Bank Bill)
Returns are for periods to 31 March 2016. Past performance is not an indication of future performance.
*Returns relate to our Accumulation (not Pension) investment options and are published after fund taxes and investment expenses, other than account-based fees.
What is ’short selling’?
Short selling (or ‘shorting’) literally involves selling something you don’t own. Here’s a hypothetical example of the basic steps involved:
- An American hedge fund manager thinks the price of ANZ shares is going down.
- The hedge fund doesn’t actually own any ANZ shares so it borrows the shares from an Australian fund, and then sells the borrowed stock on the market.
- At some point the hedge fund will need to buy back the shares and return them to the lender.
The profitability dynamics of those steps are reasonably straightforward. Let’s assume the hedge fund borrowed ANZ shares at $30. Ignoring the small amount of borrowing and transaction costs involved in establishing the short position (usually less than 1.0% p.a.), the hedge fund will profit if the price of ANZ falls below $30. However, if the price rises above $30, its short position will incur a loss.
Hedge fund activities raise the ire of some market participants who see them as unscrupulous predators exploiting the very market instability they helped create. There’s no doubt that short selling can exacerbate a market panic—sometimes leading to a ban on the practice (the most recent example being China).
We believe that the ability to short sell is fine in a normally functioning market, as it actually adds to the liquidity and efficiency of a market. After all, the root cause of the GFC was an irrational buying spree—not short selling.
Short selling is not without risks, as share prices can go up as well as down. Shorting Australian banks has indeed been a losing trade for a long time (referred to as a ‘widow maker’ in market lingo), but this hasn’t stopped hedge funds from continuing with the practice.
Why Aussie banks are perceived to be the next ‘Big Short’
The Australian Stock Exchange (ASX) discloses the amount of a company’s shares that is shorted. For our major banks the number is currently around $7 billion—close to a record high. Although it’s certainly a big number, it represents only 2% of the total value of the banks. This compares to Myer, for example, which has around 11% of its market value shorted!
Given the dearth of hedge fund managers in Australia who can short-sell, it’s reasonable to assume that most of this activity originates offshore. Anecdotally it appears that American funds are the primary source of the selling. The term ‘Big Short’ is the title of a book written by Michael Lewis which (simply and colourfully) documented the GFC through the eyes of four very successful hedge fund managers. The book was recently made into a movie of the same name.
Exactly what’s motivating the short sellers of Australian banks is open to conjecture, although the recurring themes are that Australian house prices are in a bubble, and that our household sector is highly indebted.
The following two graphs capture the essence of the short selling argument.
Graph 1: House price to income ratio
Source: Goldman Sachs Global Investment Research
Graph 2: Household debt to income
Source: Reserve Bank of Australia
Graph 1 shows asset prices growing at a rate far in excess of income growth. This has been made possible by increased borrowings as shown in graph 2.
History tells us that extreme valuations fuelled by high debt is an accident waiting to happen. In the most pessimistic commentators’ eyes, the Australian situation resembles the bubble we witnessed in America and Ireland leading up to the GFC—and we know how that ended. Given that around 50-70% of Australian bank loans are secured by housing, the implications of a housing crash are self-evident.
Other features of the Australian environment also bear an unfortunate resemblance to the American experience. In particular, there’s mounting evidence of an apartment oversupply projected to continue for the next couple of years as developers complete the current construction pipeline. Sharp falls in prices (up to 30%) are now being recorded on some apartments bought off-the-plan at the height of the boom and it seems clear that some developers and investors will lose money.
The major banks claim they have limited exposure to high risk property developers, although there’s little doubt that they have played their part in fuelling the boom. In 2014, around 40% of all housing loans written were interest-only investment loans (as distinct from loans to people who are buying a primary residence). According to our analysis, we believe at the peak, some banks were writing over 50% of new business in interest-only investment loans.
Fortunately the banking regulator (APRA) clamped down on such practices in mid-2015, limiting investment loan growth to 10%. While it’s comforting to see bank lending on a more prudent path, it is somewhat of an indictment of bank management that it has required the ‘big stick’ of the regulator to make it happen.
Not all bubbles burst; some just deflate
Forecasts of a crash in Australian house prices are not new, and of course the property market didn’t come through the GFC unscathed. Property prices will always remain vulnerable to large systemic shocks (principally recessions). However a general collapse in housing prices leading to a sharp rise in bad loans and write-offs for the banks is far from inevitable, given some mitigating factors.
Compared with the commonly referenced data in the first two graphs, graphs 3 and 4 paint a very different picture of the state of household finances. Unlike graph 2 which compares the amount of debt to annual income, graph 3 compares debt to total assets.
Based on this data, Australian households—on average—currently look far from a debt crisis, with the value of assets about four and-a-half times greater than the value of debt.
Clearly the law of averages means that there are individuals above and below the average. However, we gain some comfort from the stricter lending criteria in recent years, which should help limit borrowers from over-extending.
Graph 3: Household debt as a percentage of assets
Source: Reserve Bank of Australia
While the level of debt to assets is important, to complete the picture we also need to look at debt serviceability. That is, how onerous is it for Australian households to meet their interest payments? Graph 4 shows that that we are actually in a very comfortable position. On average, at current interest rates, only 7% of income is required to meet interest bills. On this measure, household affordability is nearly as cheap as it’s ever been.
However, this is unlikely to be the conclusion if you only considered the data in the first two graphs. Nevertheless, assets can go down in price whereas outstanding debt only falls with actual repayments so Graph 2 is not totally irrelevant. It is arguably useful in estimating the potential extent of a debt crisis rather than drawing firm conclusions on the probability of it occurring.
Graph 4: Debt servicing (housing interest payments to income)
Source: Reserve Bank of Australia
Unemployment is key
In summary, the pessimists will point to statistics which on face value look alarming, but are also potentially misleading. Along with our mining companies, Australian banks have been a favourite target for short sellers who bet that elevated house prices, fuelled by over-leveraged households, will lead to a crash.
The reality, as explained above, is that ‘on average’ Australian households have assets well in excess of debt and even if asset prices fall, the ease with which the debt can currently be serviced provides a cushion if income is maintained. And this is the crucial point—it’s all about employment.
If Australia can maintain unemployment around current levels there‘s no reason why the bubble has to burst. It can simply deflate, with a gradual decline in house prices and gradually rising incomes.
Fuelling the fire—bad corporate loans
With our major bank shares currently trading around 30% below the highs of March 2015, the short sellers appear to have the upper hand, despite the absence of a housing crash. How so?
The latest swoon in the share prices followed ANZs announcement that impairments (i.e. expected losses) on their corporate loan book would be “at least $100 million higher” than previously guided. Recent failures such as Dick Smith, Slater & Gordon, Arrium and Peabody Resources are well known to the market so an increase in impairments (particularly from a historically low level) should not come as a surprise. However, in the week following the announcement $8 billion was wiped off ANZ’s market value!
The market’s reaction reflects concerns that ANZ’s announcement is the tip of the iceberg, and talk of dividends being slashed to shore up capital is gaining traction. In our view, while a cut in dividends is possible, it is premature for anybody to be predicting that they will be “slashed”. And by no means are all of the banks equal.
During the GFC the major banks cut their dividends on average by 20%. However, to put this in context, at that time CBA’s ratio of bad debt costs to total loans was around 4.5 times higher than the level reported in their latest announcement of financial results.
So at this point it seems that the short sellers are right, but for the wrong reasons. "So what?" you might say—investment returns are best defined by numbers, not pictures.
This is not intended to be an endorsement of any of the listed securities named above for inclusion in personal portfolios. The above material reflects UniSuper’s view at a particular point in time having regard to factors specific to UniSuper and its overall investment objectives and strategies.
Past performance is not an indicator of future performance. This information is of a general nature only and may include general advice. It has been prepared without taking into account your individual objectives, financial situation or needs. UniSuper’s investment strategies will not necessarily be appropriate for other investors. Before making any decision in relation to your UniSuper membership, you should consider your personal circumstances, the relevant product disclosure statement for your membership category and whether to consult a licensed financial adviser. This information is current as at 11 April 2016.