The last post looked at the increasing margins on Australian mortgages and small business loans. On the way is another post that tries to estimate how much the banks’ own margins have been increasing. Interesting though that may be, the real problem with Australian mortgages has nothing to do with whether bank margins are or are not going up. The problem is the product itself. This post explains why.
There was an article in the the Sydney Morning Herald today which explored exactly this issue, pointing out that Australia’s “standard variable rate” mortgage, which is the most common type of mortgage in Australia, is quite an unusual type of mortgage by international standards.
Banks tend to talk about “standard variable rate” mortgages, but a better term used in the industry is “discretionary variable rate”. The problem with Australian mortgages is encapsulated in that word “discretionary”. I can clearly remember almost 15 years ago trying to explain to European and US investors who were considering buying Australian mortgage-backed securities how a discretionary variable rate mortgage worked. The conversations went something like this:
INVESTOR: So, the bank can change the interest rate whenever they like to whatever they like?
ME: Yes.
INVESTOR: Why would anyone ever accept a mortgage under those terms?
ME: Well, it’s the standard product, so people are used to it and in practice the banks tend to just change the interest rates in line with the Reserve Bank cash rate.
INVESTOR: But they don’t have to do that?
ME: No.
Why were these investors so surprised by these sorts of mortgages? It’s certainly true that in many other countries, such as the US and France, the most common type of mortgages have fixed rates, but variable rate mortgages are found all over the world too. The difference is that most of these variable rate mortgages are pegged to some kind of indicator rate that the lender cannot control. Sometimes referred to as “tracker rates”, these mortgages would specify a fixed margin (say 2%) over a benchmark rate. This benchmark may be a central bank cash rate or some other kind of short-term market rate, but the important point is from then on that margin can never change. In contrast, with Australian mortgages, variable rates move up and down with market interest rates, but banks can also tweak the margin over market rates whenever they see fit.
Last year Westpac was pilloried when it tried to use the analogy of a banana smoothie to explain why mortgage rates were rising. It may not have worked for Westpac, but the analogy can help to highlight how strange discretionary variable rate mortgages are. Imagine that the cost of bananas goes up due to a cyclone-induced banana shortage. It may well be that the price of banana smoothies goes up (although it may also be that café owners take a portfolio view of their business, value their customers and absorb a bit of margin compression on their smoothies, but that’s another story). What certainly does not happen is that café owners go around to everyone who has bought a smoothie in the last year, explain to them that bananas are now more expensive and demand that their customer pays a bit more now for last year’s smoothie.
That is essentially what happens with discretionary rate mortgages. You might have taken out a mortgage a few months ago after doing extensive research comparing interest rates and deciding that the best value mortgage you could find was from the Commonwealth Bank as it was 0.1% cheaper than the next best offer (this may or may not have actually been the case). So far Commonwealth Bank is the only bank to have hiked their mortgage rate by 0.2% more than the Reserve Bank and now your “cheap” mortgage is 0.1% more expensive than the bank you turned down. So much for shopping around! Banks may argue that you are free to change to another bank if you are unhappy (although you can expect exit fees, particularly if you received any kind of rate or fee reduction when you first took on the loan). This does not change the fact that it is a rather unusual product that allows the seller to increase their margins after they have done the deal.
This hypothetical example highlights one of the real problems with the discretionary variable rate mortgage. It is inherently anti-competitive. There is little point shopping around for the cheapest mortgage when after next week it may not be the cheapest any more and you are locked in for 25 years. Is it any wonder that most people shrug their shoulders, say that the banks are all as bad as each other, hold their noses and just pick one almost at random?
There is another problem with discretionary variable rate mortgages, as one Mule reader pointed out in an email. It has the surprising effect of creating some credit risk for the borrower. Normally, depositors are exposed to the risk that the bank will fail, while banks are exposed to the risk that the borrower will fail. But, if you take out a discretionary variable rate mortgages, you may end up paying more if the credit quality of the lender deteriorates. The Herald article gave this hypothetical scenario:
Suppose one of our banks got downgraded from a AA to B. What would happen at the moment is they would just increase the margin on their mortgage rates to cover the extra costs they would face, whereas that risk should fall on the management and the shareholders.
But this sort of thing actually has happened! Many of the non-bank lenders like RAMS got into trouble during the global financial crisis and found funding through securitisation difficult, if not downright impossible. Some collapsed or turned to banks for support, but all of them suffered fast rising costs. Many borrowers who took out mortgages with these lenders saw their interest rates go up as a result. Some were able to refinance their mortgages with another lender, but those struggling the most to pay the higher interest rates would also be the ones least able to get refinancing approved.
In my view, abolishing discretionary variable rate mortgages, though unlikely to happen, would be a good thing for the Australian market. There’s certainly no guarantee that margins would drop. But it would change the stakes for banks considering raising rates to preserve their margins. Rather than being able simply to recoup that margin from their existing mortgage book, they would have to seriously consider the impact the move would have on new business, because it would only be new loans that would be paying the higher margins.
Possibly Related Posts (automatically generated):
- Cash rates and mortgage rates (4 May 2010)
- Where does the money go? (31 December 2010)
- Banks, banks, banks (5 November 2010)
- Bank funding costs (16 November 2010)
I would love to see the customer stampede if one bank were bold enough to go for market share by offering a 30-year fixed rate / zero fee mortgage. That kind of revolution can happen even in markets with only a few players. When the first telco offered cap plans for mobiles, the rest moved to match it very quickly, even though it sucked hundreds of millions in profit out of the sector overnight. Now, we don’t expect anything less. Has any bank offered this?
Steve: good point! No-one has offered a product like that here yet. I won’t hold my breath, but it would be good to see. To work, I think they’d have to go the whole way in replicating the US model and also make the break cost for early termination zero (maybe that’s what you meant). Otherwise, early break costs on a 30 year mortgage could be pretty scary if rates had fallen. That then presents a bit of a chicken and egg problem. Banks in the US use long-dated interest rate options to hedge the risk of people refinancing their mortgages when interest rates fall. There is no real market in long-dated options here in Australia. Of course, the main reason that market is active in the US is precisely because of the 30 year mortgage product. Which came first??
Discretionary mortgage rates have surely encouraged the banks to lend money at excessively low rates during the good times, or when competition is fierce, since they know that they can always raise them later. As you say, they can push some of their risks onto the borrowers.
If you think about it, every mortgage in Australia (even the “fixed” ones, as they’re rarely fixed for more than a few years) is a “discretionary reset’ ARM, which are one of the categories of mortgage that went pear-shaped in the US…
Danny you may be right there, although a big part of the problem with US ARMs was that they typically had very low teaser rates to begin with and, although benchmarked to benchmark rates, the margin was contracted to jump up significantly after some initial period (usually in the 1-2 year range). The size of some of these jumps were so large, that Australian banks could never dream of getting away with stepping up variable rates that quickly.
Food for thought, Stubborn…
But there’s something I don’t understand, in Peter Martin’s piece, he says:
“The chief of the Bankers Association, Steven Munchenberg, is also sympathetic. He and his members are looking at the Greens’ proposal for each bank to offer such mortgage accounts as an option.”
If Adam Bandt’s proposal establishes a kind of pegging with a benchmark, shouldn’t the banks be howling and wailing?
Magpie: I can only speculate (which I am happy to do). Given all the media and political flak banks get, they don’t really get as much leeway with discretionary variable rate as they might like. Perhaps it would just be easier to have tracker-rate style mortgages, but the trade-off might be somewhat higher margins. If a tracker mortgage is offered, but as a separate product with a different (i.e. higher) rate, that might be an attractive way to go for the banks. In any event, it would not be as painful for the banks as simply saying that all existing mortgages must only move in line with the RBA cash rate from this moment on.
Definitely food for thought – a 30 year fix, now wouldn’t that be handy!
Out of interest, what happens overseas to a bank who has offered a tracker mortgages based on a slim differential on an index that was high when the mortgage was sold, but then falls markedly during the life of the loan?
For example in the UK most variable mortgages are pegged to the bank of England base rate, which has fallen more than 5% during the GFC to a present low of 0.5%. I have friends who took out 5 year tracker mortgages in 2007 when rates were at their highest on very small base rate differentials (and have even heard stories of some mortgages being sold on negative differentials during that time!) presumably because the banks could actually obtain funding at a significantly lower cost than base rate at the time.
Are the banks now losing money on these products or do they have some way of protecting themselves from this situation?
Another disadvantage of discretionary rates in comparison to trackers is that they weaken the monetary policy power of the central bank in managing the economy — banks may or may not pass on adjustments to the official cash rate to households with mortgages. Of course, fixed rate mortgages are worse in this respect.
Mark L
Pete: I don’t know why some UK banks would have offered such tight margins on their tracker mortgages (maybe there are fees as well, or they hope to make the money from other products like credit cards), but since their funding costs will have fallen along with the BoE base rate, they should not really be any worse off with these tight margins now than they were when rates were higher. Where they would see their profits squeezed would be on low/zero interest rate deposits where falls in the base rate do erode their margins.
I worked at SBNSW in 1993-4 and they had a novel 10 yr fixed rate mortgage product with no exit fees. Of course there was a premium to pay for the option (which there is in the US but built into the market and disguised by low overall interest rates). I believe from a marketing point of view the target was to get the rate under 10% so it was probably something like 9.90%. It wasn’t a dog but I don’t think the bank continued the product much after it’s initial release.
The problem with such a product is there is a lot of “non-rational” (to a banker!) behaviour to take into account and this needs to be modelled for the purposes of hedging (even with a liquid 10 yr option market the delta is not 1 due to non-rational behaviour, in practice in Oz you hedge with 10yr futures anyway) and hence pricing.
In the US there is a liquid market in derivatives on long-dated RMS that incorporate this behaviour implictly so hedging can be done directly in this market. This means theoretically at least the US market should be a lot more efficient than the Australian market and the risk premium for the option should be cheaper.
So there is more to this than just introducing long-dated fixed products without exit fees. They have to be competitive as well and this requires a liquid traded market in derivatives on RMS securities.
This seems to be a recurring theme in the Australian financial industry. I wonder how much is tied to the regulatory environment. Banks are effectively punished for taking on risks, by APRA in particular. Mortgages have been like this for a lot longer than APRA has been around but other similar trends (like the move to defined contribution superannuation) line up with the introduction of APRA fairly well.
On a related subject:
Mortgage stress to intensify if rates jump. SMH. November 12, 2010
http://www.smh.com.au/business/mortgage-stress-to-intensify-if-rates-jump-20101112-17qwd.html
Tense, nervous, mortgaged. Jessica Irvine and Phillip Coorey. SMH.
November 13, 2010
http://www.smh.com.au/business/tense-nervous-mortgaged-20101112-17r3j.html
Matt the move from defined benefit to defined contribution super by corporates was an effort for them to take less risk in their super liabilities, but my understanding is that that risk reduction was something that appealed to companies (after all, it’s not a risk that generates them any return) rather something pushed on them by APRA.
Magpie there should be no doubt that every rise in interest rates increases mortgage stress. After all, there will always be some people on the limits of their abilities to service debt before the rate hike, then the move pushes them over the edge. Having said that, the survey figures quoted in the SMH article should be taken with a grain of salt. Respondents were told by how much typical mortgage repayments would increase if rates went up by 1.5% and asked what they would have to do if that happened. Apart from the fact that rates have since gone up around 0.4% and many think that the hikes are done for now, no-one would like to see their mortgage repayments go up, so I think that there is a real chance that respondents may exaggerate the extent of their response to rate rises. I could be wrong of course, but these sorts of “what would you do if…” surveys are notoriously difficult to get meaning ful figures from.
I think you are right, but what I meant to say was that the Australian Financial industry seems organised to reduce the amount of risk taken on by these large companies. This is a good thing, except when it is not. I have been a long time out of the industry, but I understand that the more risky the investments/mortgage, the more reserves the regulators insists on, so the less money you make on the risky investment/mortgage, effectively making risky investments unattractive. At some point regulatory risk aversion results in large companies all pushing risk to their customers. No-one can get a competitive advantage offering the riskier products, since it is not their decision what the consequences/tradeoffs/potential profits are, they have to play whatever game the regulators insist on.
Matt: I’d agree with that. Furthermore, many retail customers may not like risk, but dislike paying for risk even more as Zebra points out. Given the choice of a less risky mortgage that may have an initial interest rate that’s a little higher, many borrowers will take the lower rate regardless of how much extra risk in the future that exposes them to.
Stubborn,
“Having said that, the survey figures quoted in the SMH article should be taken with a grain of salt. (…) I think that there is a real chance that respondents may exaggerate the extent of their response to rate rises.”
That’s a good general point, without any doubt. Taken in isolation, it would have settled the argument.
However, I think you are not taking into account additional information, which, if considered in conjunction with the SMH notes, lend those notes a lot more credibility.
Have a look at “4130.0 – Housing Occupancy and Costs, 2007-08” (ABS). By the way, you might remember I’ve mentioned that release in the Stable, some time ago.
You’ll find that by 2008, according to Table 10.9 (Summary of Findings), 31.6% of households were already employing over 25% of their weekly gross income to service their loans.
Further, included in that 31.6% were:
(1) 22.1% of all households with a mortgage were paying over 30% of their weekly gross income, which qualified them as being in a state of “housing stress” (from Table 5).
(2) 7.8% of households with a mortgage, for which housing costs represented more than 50% gross household income (from Table 10.9).
(3) There are some methodological issues (which I will not touch here) that are highly relevant to those results and make them, if anything, too conservative.
I’m sure you’ll have noticed (making allowance for the time elapsed and the fact that in both cases we are talking about survey data) those figures follow quite closely the figures mentioned in the notes:
“The modelling by the National Centre for Social and Economic Modelling found more than a third of Sydney households with mortgages, or 35.5 per cent, were experiencing ‘mortgage pressure’, where repayments took more than 30 per cent of disposable income – the highest of any capital city and above the national average of 27.7 per cent” (Irvine and Coorey).
“The figures also reveal the number of people under extreme mortgage pressure – spending more than 50 per cent of their income on repayments. In Sydney 11.9 per cent of households are under such pressure, compared to the national average of 8.4 per cent” (I & C).
There are a few things that make it difficult to draw immediate conclusions, however:
(1) The data reported by the ABS was “collected (…) over the period August 2007 to June 2008” (Explanatory Notes).
(2) During that period average variable standard housing loan rates were 8.85% (Table F5 Indicator lending rates. RBA).
(3) As of Oct-2010, variable standard housing loan rates were 7.40%.
(5) I am not sure whether these RBA figures include or not the interest rate differential (spread) you mentioned in “Banks, banks, banks”. If they do not, which seems possible, we’re already around the same level of effective rates as in 2008, contrary to what the SMH notes say.
(6) The ABS definition of “housing stress” does not consider other debts (as pointed out in the notes).
(7) It’s quite likely households weekly incomes have fallen, given the level of underemployment.
(8) There is empirical evidence (which I will keep for myself for a little longer) that bankrupticies have been on a moderate but sustained increase.
Just my two cents.
Stubborn
By the way, the bankruptcy data I am talking about does not discriminate between housing debts from other types of debt (personal and commercial).
Hence, my new found interest in commercial loans.
Pingback: Bank funding costs
just been slugged $16 500 by ING for break costs on a $318 000 loan
Belinda 5% is hefty! Was it a fixed rate loan? Mind you, with interest rates on the rise, I would not have thought even a fixed rate loan would get trigger much in the way of break costs.