Competition in the Financial System
Peter Harris presented a speech to the Committee for Economic Development of Australia (CEDA) on 26 February 2018 in Melbourne.
Download the speech
Read the speech
Banks and insurance companies have special status in any developed economy, and deservedly so.
In Australia today, our money is safely held, the costs of doing so are reasonably well managed, and technology to allow us access to our money is adapted and applied reasonably rapidly.
And banks and insurers do try to apply different technology strategies in particular to attract different classes of customer.
Collectively, they manage risks and facilitate transactions that – without the high level of confidence created by a network of dependable parties – none of us individually would be able to manage effectively… bitcoin notwithstanding.
So far, so good.
With this special status comes both strong support via regulation and an unwritten expectation of social responsibility.
And these involve in turn the formation of understandings and relationships and activities up to and including price-setting that are unique amongst private sector markets.
The Productivity Commission can be an unwelcome presence in unusual environments like these. There is a lot of money, pride and power tied up in finance. And a lot of system awareness for us to pick up in a short time.
Yet we believe there is value in the occasional analysis from the ground up, taking nothing for granted, as is our special skill. It is a practice not much seen today. We welcome the chance to undertake this task.
In our modern guise i.e. since 1998 the Productivity Commission has never been asked to look at the finance sector as a whole.
We have looked at aspects of it – superannuation, small business finance, bankruptcy and some taxation angles - in recent years. But until David Murray recommended in his Financial System Inquiry that we look at the sector’s overall competitiveness, the finance sector has not seen - nor I expect known - much of us.
Finance is not alone. The core policy aspects of health or education are equally important elements of industry structure in Australia, and they too have not seen much of us, although we have tried to offer some analysis in last year’s Shifting the Dial 5-yearly Report.
Together these three add up to more than 20% of the economy….and expanding, in government influence as well as private investment.
We received good assistance from the regulators in the course of this novel inquiry – APRA, the RBA and ASIC.
Our report does propose important but subtle changes to how they work with each other and with the industry.
This was not because we found them to be lacking in the crucial ingredients for good regulators: high levels of competence and a willingness to consider alternatives. Far from it.
But rather because the degree of competition in the finance sector’s principal consumer markets is substantially a factor created by regulation and regulators.
And consequent upon that is perhaps the most important underlying theme in the draft report: that without proactive regulator thinking on competition, we should not expect anything other than very occasional bouts of fierce competition in the principal banking and insurance markets.
Put another way, serious rivalry – particularly in price – will for long periods of time tend to be dormant.
In saying that, we are very conscious that it might suggest we have given up at the draft Report stage. So I feel obliged to say we have not. But if we are to be effective in better policy design, we first have to be realistic.
Reality appears to work roughly like this. Fierce competition brings with it the risk of instability. And instability is not acceptable over the medium term to the regulatory structure, acting in the interests of depositors (i.e. us) and the broader economy.
Thus regulators apply controls on the degree of competition faced by many market participants. And since the GFC, regulators’ ability to and willingness to intervene has risen, aided by international standards to which we in Australia as a capital importer necessarily need to broadly adhere.
What this means in sum is that regulators must be at the heart of our Report.
And our recommendations affecting them are both powerful and yet focused on intangibles: not on rank failures, but rather the nuances and subtleties in regulatory arrangements and their impact on consumers.
I will talk about a few of these today.
While regulators were helpful, it was inevitable that, having been subject to so many inquiries in recent times, parts of the industry more broadly reacted with weary resignation to yet another one – ours.
Plus knowing that data is potent in the hands of quality analysts, we understood too the deep reluctance we encountered occasionally to the sharing of data. Disappointing as it nevertheless was.
And finally, in a sector where concepts like confidence are crucial, we recognise that it can be problematic for others to explain to our analytical satisfaction how intentions were translated into actions. We believe we have given this art rather than science due weight, when drawing conclusions.
But we did not back off, for all that.
We are asked by our Terms of Reference and the legislative statute that guides us to make judgments in areas where, in this case, competition appears to be lacking and where responses would be both cost effective and necessary.
So in a draft Report, we take a deep breath; we note the limitations of the evidence; we risk the criticism that at times must follow; but we nevertheless persist and use the draft to provoke debate.
We do it in order to encourage that debate of ideas. We hope to see more light emerge from such debates as we move into the public hearing phase, this week.
I mentioned a moment ago the potency of access to data.
When the topic is competition, being able to demonstrate whether or not parties hold market power, and if so the purposes for which they are using it, is a data-driven exercise if done properly.
We think we have demonstrated that there has been an exercise of market power in key product lines - deposits and home loans for example. The larger elements of the banking industry have shown an ability to maintain comfortable margins despite the shock of the GFC, followed by an era of unpredictably rapid falls in official rates; and yet to also keep market shares relatively stable. This is documented in the report.
I don’t think I am being unfair when I say the default position amongst data holders in this industry is set against transparency.
But we were genuinely surprised to find that they either do not hold data at all on some important aspects of decision-making, or for another reason could not supply them.
For example, the cost of mortgage brokers is quite high, $2300 for the average loan of $369,000, plus a trailing commission (more on that later) of $665. Other analysts have suggested higher numbers than these in high-priced locations but we will stick with national averages.
More than $2.4 billion is now paid annually for these services.
Some in the broking industry want to know why there is suddenly attention being paid to commissions. The sum I just cited, as a large apparent addition to industry costs since the mid 90s, by itself suggests a public analysis of why it is so large might be in order.
Wealth advisers, much derided though they are, seem to handle investments of the same order for much lower costs.
Moreover, the sum becomes problematic when it is also suggested that customers aren’t burdened by this as they don’t pay these costs. Which is a comment surely made for twitter - since anyone with a slight amount of commonsense knows that somewhere in any product purchase it is only a customer or a shareholder who could be paying this charge, unless offsetting costs have been stripped out.
Shareholders returns are pretty constant, so we would have liked to unpack that cost question a little, to see if the price was supported by cost savings.
With the data provided by banks, this proved to be near impossible. For smaller banks, we were able to develop some estimates of the branch costs they would potentially face, without broker assistance. But we received insufficient information from most (not all) banks, and so could not create a clear picture.
Thus we can’t say whether there has been a net improvement in efficiency, even as a large sum in commissions has been added to industry costs.
We have also shown in the report that brokers do produce slightly better rates for their clients than going in to the bank branch. But that benefit for consumers has been declining since the GFC. It would have been valuable to put the cost:benefit side by side.
Making the presumption – accurately, as it turns out – that we would find strong indications of the exercise of market power, some commentators had hinted in advance of the report that we might have major structural solutions to propose, like forcing break-ups of vertically integrated operations.
Our assessment is that the case for this, on the information we have to date, is not sustainable - for two reasons, primarily:
First, there are efficiencies to be found with vertical integration and they are observable.
Second, the solution needs to fit the problem. The problems that have been observable with integration are ones of real or claimed bad behaviour. It is also evident that there are poor performance incentives.
When direct solutions to this are available - and they are, we recommend some - forced divestment should be a last resort, rather than a first one.
Of course, if the necessary solutions prove commercially unpalatable, institutions themselves may then choose to divest.
Grouped under the usual ways of improving competition, in the draft Report you can see recommendations for:
- better information to consumers
- better transparency on the part of regulators and governments
- removal of conflicts of interest; and
- lower barriers to entry for both suppliers of product (e.g. new banks) and suppliers of services (e.g. wealth advisers might compete with mortgage brokers)
- an access regime for a key piece of national infrastructure
It’s also strongly oriented towards taking advantage of better data in a digital world, as you would expect of the PC after our Data Availability and Use report last year.
Perhaps the most important data-based improvement lies with the proposed publication by ASIC of actual home loan rates recently negotiated, in as close to real time as possible.
Until recently, it would have been radical to suggest that banks could and should publish their actual median home loan rates in near to real time, across classes of customer and location. But today, it is not.
On its face, in fact, it is pretty absurd that customers still have to guess at what is a competitive rate for their new home loan might be before they commence negotiations. And try as we might, we could find no sound reason why this was so.
In no other industry is the price of a major purchase so apparently shrouded in smoke and shadow, featuring concepts like the standard variable rate, or the comparison rate.
Perhaps these notional benchmarks were once well-intentioned substitutes for reality, when comprehensive information was not gathered and could not be published quickly.
But not today.
The sector has the ability to keep the market fully informed of price differences when they emerge, even if they are only the product of weak competition. Today, we have digital data. It’s cheap, it’s already assembled inside each financial institution and it’s not radical any more to think that such prices can be published to add to consumers’ knowledge.
In a highly competitive market, some institution would probably seize the initiative and launch this themselves, for the positive brand recognition value alone.
Because of the opaque nature of pricing, many consumers feel the need for brokers.
The typical home loan is:
- a major commitment
- requires a sound ability to judge risk
- has a price that you do not know until you have spent serious time on the transaction
- may involve the obligatory purchase of other services (Lenders Mortgage Insurance, for example)
- is a means to another purchase, rather than an end in itself.
This is very complicated. And mortgage brokers and mortgage securitisation were certainly a godsend for a period from the mid 1990s.
They broke down both knowledge and capability barriers: knowledge of the system and where the opportunities lay; and capability to negotiate a deal.
No wonder they thrived and consumers benefited. Prices fell and service quality rose. This was the last time we had fierce competition in the largest loan markets.
But since late last decade, this revolution has been slowly but surely absorbed into the establishment. About 70% of the broker channel is now bank-owned or controlled.
We are not arguing that the better use of real time actual price data would so simplify things for consumers as to make brokers redundant. There are many other aspects to a loan, beyond price, that still may intimidate a first-time home owner. But it would certainly offer a choice for them as inexperienced consumers.
And competitive markets thrive on choice.
There is also a benefit in publication of actual prices for existing mortgagees. This might help them to recognise when loyalty isn’t being rewarded, and seek a better deal.
Genuine price data will also complement the duty of care we propose, to ensure a bank-owned broker acts in the customer’s best interest.
As I noted a moment ago, on our estimates banks’ broking aggregators have about 70% of the home loan broking business.
We noted in the draft Report that at the time banks decided to move in to buy up brokerages – a legitimate commercial decision, we do not find a case on competition grounds to oppose integration, nor to this point has the ACCC - they might have made a conscious decision to address what seem obvious conflicts of interest.
For example, a consumer expects the broker will negotiate for him or her, and not favour the bank. Ownership surely complicates this.
Experience with similar conflicts of interest in wealth management and the disastrous effect they had on some aspects of banks’ public standing might have further encouraged such thoughts.
But that did not happen.
Brokerages’ Approved Product Lists may thus give the illusion of choice, with an array of branded products sourced from a single lender, or white-labelled to tie customer loyalty to that brokerage.
And not all brokers at a particular brokerage need even be accredited to offer all the products that the brokerage advertises. It’s an extreme scenario, perhaps, but your broker may only be accredited to advise on house products, despite what is on the glossy paper.
Consequent or not, in-house products appear to dominate disproportionately the outcomes for borrowers who use bank-owned aggregators. In 2015, the Commonwealth Bank had 21% overall market share in the broker channel but a 37% market share via its aggregator Aussie Home Loans.
Beyond the incentives that are influenced by ownership are the incentives faced by an individual broker.
Despite some recently announced industry changes to parts of the commission payment structure, commission earned by brokers remains far from aligned with the interests of the customer.
Trailing commissions are an example of that. These are only paid while a customer remains with a loan. They are worth $1bn pa. There is nothing immaterial about them.
The industry itself has said that trailing commissions are designed to reduce churn and manage customers on behalf of banks.
Despite the hint to the contrary, we do actually understand quite well why it might be in a bank’s interest and a broker’s interest to jointly limit churn.
But not the customer’s interest – who (the data is surprisingly unavailable, as noted earlier) is most probably paying for the service.
Given the unhappy experience with misaligned incentives in wealth management, being able to substantiate the assurance that a broker is acting in the customer’s best interest would seem to be pretty desirable today.
As the Productivity Commission, we would prefer that, rather than regulation, the banks imposed this duty on their brokers, perhaps via contract; and promoted it, to the benefit of their reputations.
But we have no power to recommend what banks do for themselves, so we have instead a draft Report that proposes regulation.
For independent brokers, the case is a little more nuanced:
- first, the data does suggest that brokers as a group are still achieving slightly better (but less so in recent years) results for customers than simply going in to the branch
- second, the wealth advisory industry since the FoFA reforms has experienced increasing market concentration, and we would not want to add to market concentration for this service until we can determine better whether there is cause and effect.
Thus the draft report limits the call to regulate to bank-owned brokers.
But we are also considering applying additional competition across all brokers, as another possible way of reducing cost to consumers, by allowing licensed financial service providers such as financial planners to offer home loans as a product.
There appears to be no sound reason why qualified advisers are presently not allowed to deal in this product, but they are not.
We are seeking further input on this, before finalising a view.
We also found some possibly good news for small and medium enterprises and the banks that tend to lend to them.
Australia has been quite conservative in its application of risk weightings to small business lending, with a strong leaning towards a home mortgage as the preferred security basis for a bank loan.
Other countries have, without threatening their system stability, taken up Basel standards that diversify SME risk weightings, lowering them in some cases. We have recommended APRA take account of this.
What this means is that banks may require in future less capital to hold against small business loans that are not secured by a house mortgage; and could expand their appetite for such loans, within their current capital allocation.
We believe APRA now has this under active consideration. Our final report will provide an update on the Recommendation.
We also have proposed reforms to payments systems where current practice may prevent consumers and merchants from taking the lowest cost option for making in-store payments.
Small businesses are least able to manage this conflict and the changes proposed could reduce cost and inconvenience for them.
Moving from the important and worthy to the controversial, the Finding we made on the 4 Pillars policy made quite a few headlines.
We called it redundant policy. What that means is that if it ever once had a role in sustaining competition, that day is long past. Thus it is redundant.
While some have gone so far as to suggest the radicals in the PC have lost their minds and put competition under threat, let me show you what the Wallis Inquiry in 1997 actually said, and judge for yourself how radical we are, twenty years later:
Competition concerns appear to have been the original justification for the ‘six pillars’ policy. However, as this Report demonstrates, the financial system is dynamic and the pace of change is likely to accelerate, not slow down. Thus, any static policy may become outdated.
Also, as discussed above, the Inquiry has taken the view that the financial system should be subject to the same set of competition rules as the rest of the economy – namely, those contained in the Trade Practices Act and administered by the ACCC – and that no other competition regimes should be applied to the sector.
It follows that on competition grounds the Inquiry does not support continuation of the ‘six pillars’ policy, or a modified version of it.
This position should not be interpreted as representing a view on the desirability or otherwise of mergers among any of the ‘pillars’. Rather, the position simply states that the ACCC should assess the competition implications of any such proposal. It should do this in accordance with the merits of the proposal at the time it is made.
‘Redundant’ is a state of relevance, or lack thereof. It really doesn’t matter which politician swears that he will preserve this policy, it is and will for competition purposes be redundant regardless of such pledges.
Thus our point is not that removal of the 4 Pillars policy would encourage competition. Like Wallis, we would not argue that three or two is better than four (or six, covering insurers too, as it was back in the 1990 decision that gave birth to this tenet of policy faith).
The problem with such policies is the comfort that they induce. In this case, the false hope that banks must be competing because we have and will always have at least 4 of them.
We know policymakers aren’t fooled by this (and neither are banks). So what’s the harm in keeping it?
In a broader national welfare and efficiency analysis, which the PC’s legislation invites us to consider whenever we receive a Terms of Reference, it is probably shareholders who have most reason to complain.
By the by, that’s just about all of us via our superannuation funds.
The 4 Pillars policy removes an important market discipline from these four major businesses, should management make serious errors of judgment.
As David Murray has lately observed, even if the 4 Pillars policy was put through the shredder, there are still direct legislative powers that limit ownership in any licenced deposit-taking business.
We recognised this. We even have a Recommendation on it.
But in our analysis, far from being a reason to keep the 4 Pillars, lifting the allowable level of ownership in any bank – not just new banks – might introduce some market discipline on misadventures in say US home lenders or UK banks or foreign exchange markets …over the past decade or two, there have been a few of them.
As I said earlier, it’s an issue really for shareholders, thus in this competition analysis we stop there, and leave it to them.
Lest we forget it, the ACCC will always have a say in any attempt at change of ownership that prospectively substantially reduces competition.
Thus from a competition perspective, we are inclined to consign the 4 Pillars policy to the shredder.
The other controversial proposal we made was in regard to the exercise of macro prudential intervention.
Just to repeat, we were asked to undertake an analysis of competition and influences on it. Not a review of prudential success. Having both is desirable, and we can.
Macro prudential intervention – directions to banks from the regulator, APRA, that is charged with maintaining overall system stability - is pretty novel stuff, here and around the world.
Novelty has been most evident since the GFC, but in fact the prudential regulator will still have an effect on competition in a future time when shifts in the official cash rate are again common. So the impact of such interventions was clearly within scope for us.
Broadly stated, risk is what will motivate APRA to intervene. Such was the case in 2014 and again in 2017.
And yet risk-taking is at the heart of competition. And, dare I say it, innovation. And thus to productivity.
Our focus is on the quality and availability of the pre-implementation analysis of the potential damage to competition. We do not propose that APRA would be fettered in subsequently doing whatever it chooses. We have confidence in its ability to consider the matters raised by any improved analysis.
And again we are not being terribly radical. Strengthening the consideration of competition beyond the present general legislative guidance to ‘have regard to’ effects on competition has relatively recently been the subject of alteration to the remit of the Bank of England’s Prudential Regulation Authority. The UK acknowledged, in doing so, the need for more proactive approach to competition. A subtle change, such as we too propose; but a powerful one, for all that.
The lack of publicly available pre-implementation competition analysis and the way the banks subsequently responded to these interventions is outlined in our draft Report. In 2017, there were price rises in some market segments including pre-existing loans, along with sought-after reductions in the future rate of flow of funds to that sector. In the language of the sector, the back-book was re-priced.
Those price rises - as these were investor loans - will be deductible in part and so a cost to taxpayers.
I’m sure someone will want to re-cut our estimate of that cost, up to $500m. Despite that, the point will almost certainly remain undiminished: there was a public cost as well as a private cost to borrowers, and these should both be accounted for in any analysis of the competition effects before the action is taken.
And this is only one aspect of a preferable competition analysis.
Fortunately, and quite desirably, we believe we do not need to change APRA’s legislative remit as the UK did in order to consider the impact on competition as it designs its intervention. What we propose is adding independence of the source of advice to it, while not altering the decision-maker. This is a subtlety that may have been missed by some commentators.
Our proposal is that, in the most important forum dealing with these matters – the Council of Financial Regulators - the competition champion is not a decision-maker. Rather it is a source of advice designated by the Treasurer, to be satisfied that competition is being considered in-depth.
And we propose to add the publication of Minutes from the discussion at the Council as the intervention takes effect, to enable greater clarity about both the target and the chosen means for the intervention.
This too would seem to be a desirable outcome for both competition and stability, and may assist in reinforcing the objective.
One last thing on macro prudential intervention, before I leave the topic.
One or two people we have encountered since the draft Report was published have asked how the intervention affects competition. The thought is that there may be a theoretical case for transparency and for considering the cost to taxpayers, but is this really about competition?
For us, the connection to competition seemed self-evident but we may need to expand upon it in the final report. For now, let me say there appear to be two aspects to this:
- first, the intervention enabled notional competitors to act in concert, and provided them with a high level of awareness of others’ inability to compete, which is a direct reduction in competition
- it appears that we may be training an industry to find profit in a regulator’s intervention where banks’ costs have not risen ie quite a different circumstance to when the cash rate rises
- second, we may be in danger of operating a one-way ratchet on prices
- to the extent that regulated businesses lift prices to achieve macro prudential outcomes, there appears to be nothing in the regulators’ tool box to see such premiums fall when the risk-taking passes
- in a strongly competitive market, we could comfort ourselves that premiums will be competed away over time. In Australia, the market power held by the big regulated parties does not indicate that this is likely.
As I said at the outset, we welcome this chance to examine the competitiveness of the finance sector, it is a rare chance to scrutinise policy in a major part of the economy whose services we all depend upon.
We think the draft Report offers good opportunity for further debate about matters that rarely make it into public policy forums.
To that end, we are seeking submissions by 20 March.
Today’s comments are designed to encourage that participation.