Tuesday, April 2, 2013

Would it be costly to require banks to raise equity to 30 percent of total assets?

In their recently published book, 'The Banker's New Clothes', Anat Admati and Martin Hellwig make a strong case that in order to reduce the risk of insolvency in major financial institutions, shareholders should be required to fund their lending and other investments to a much greater extent.

bookjacketThe authors argue that government regulation to reduce the risk of insolvency of major financial firms is desirable because failure of such firms has adverse effects that are analogous to those that can arise from accidents in nuclear power plants. When I discussed that analogy in an earlier post, I accepted (somewhat reluctantly) that it is appropriate. As a result of the interconnectedness of financial markets, it would probably not be possible to avoid major economic disruption if large financial institutions were allowed to fail when they became insolvent. That makes it desirable to find the least cost way of regulating them to make it less likely that they will become insolvent. Governments are thus presented with problems that are similar to those involved in regulating the nuclear power industry to reduce the risk that serious nuclear accidents will occur.

Admati and Hellwig suggest that the best way to reduce the risk of insolvency of major financial institutions is to require them to raise shareholder equity from current levels (which under Basel III can apparently still be as low as 3 percent of total assets) to 20-30 percent of total assets. The higher ratio of shareholder equity to total bank assets would provide greater scope for any future fall in the value of bank assets to be accommodated without insolvency.

The authors suggest that requiring banks to rely more on equity funding would impose little, if any, cost to society. In this post I want to focus specifically on the reasons they give for that view. I encourage readers who are interested in a broader discussion of this important book to read John Cochrane's review.

The authors argue that requiring banks to rely more on equity funding would impose little cost on society because it would offset the bias in favour of borrowing provided by government guarantees and tax systems. Banks and their creditors benefit from explicit guarantees to protect depositors as well as implicit guarantees associated with the 'too big to fail' concept. These guarantees enable banks to borrow on more favourable terms than would otherwise be possible. Tax systems tend to favour borrowing because they make interest paid a tax deductible expense.  (The dividend imputation system in Australia reduces this bias to some extent but, as acknowledged by the Henry Tax Review, there is still a bias in favour of foreign borrowing and Australian banks rely heavily on this source of funds.)

The authors point out that equity ratios of banks were generally much higher in the 19th century, prior to the existence of government guarantees.  In the US, until the middle of the 19th century, equity levels around 40-50 percent of banks' total assets were typical and early in the 20th century it was still common for banks to have equity of around 25 percent. The picture seems to have been broadly similar in Australia. Data presented in an article by Charles Hickson and John Turner shows (apparently) that the average equity to deposit ratio of Australian banks declined from around 60 percent in the 1860s to around 20 percent in 1892. The subsequent depression would presumably have substantially depleted the equity of those banks that managed to remain in business. Adam Creighton, a journalist, implies that the surviving banks re-built their capital ratios following the depression, so that a century ago they maintained capital ratios of between 15 per cent and 20 per cent. (See: 'Time to Force the Big Banks to Hold More Capital', 'The Australian', 23 November, 2012.)

Admati and Hellwig point out that the proposed increase in bank equity would not interfere with core banking functions of accepting deposits and making loans. Given the current structure of balance sheets, the increase in equity levels would tend to displace additional borrowing from sources such as money market funds rather than bank deposits.

The authors point out that bankers' claims that equity is more costly than debt are flawed because they don't take account of the effect of increased equity in reducing the risk of bank failure and thus reducing the rate of return required by shareholders. Equity only seems costly because government guarantees provide an implicit subsidy on debt. The increase in equity could be accomplished without significantly disadvantaging existing shareholders by requiring banks to retain earnings rather than pay dividends, until equity levels have reached the minimum level.  

I am normally sceptical of claims that governments can improve matters when they attempt to offset the adverse effects of previous interventions by adding a further layer of regulation. It seems, however, that Anat Admati and Martin Hellwig have found an instance where the theory of second best provides a valid guide to policy action. There are strong grounds to argue that if governments cannot credibly bring the 'too big to fail' policy to an end, they should take decisive action to offset the effects that policy has had in encouraging banks to become more fragile.  In my view the authors' proposals deserve strong support.


Anonymous said...

Hi Winton.

This is an interesting post, and with interesting links as well.

First, I include some words from an earlier comment of mine:

"As far as I can work out our banks total funding sources might be analysed as long term external 25%; short external 20%; shareholders 7% - leaving depositor funds 48%. Of that 48% I believe roughly half is in the form of fixed term deposits, with terms of up to two years.

You made no comment on that rough analysis at the time, but I've since seen further (again rough) confirmation for those figures.

So now your post seems to favour lifting 'shareholder equity' to maybe 30% - from where it sits, at maybe 7%? That is a quite significant increase. That is not 'flick of the wrist' stuff.

Or maybe it's a problem of terminology; where I read a particular term in one way, but you may mean it in another?

Switching from a discussion of 'equity' per se to 'equity to deposit ratio', and even the simple term 'insolvency' - which to me means an inability to pay one's debts as they fall due, something not necessarily equivalent to bankruptcy - where liabilities exceed assets, which itself may mean 'failure' as you've used it.

Maybe your 30% refers to the above-mentioned 'equity to deposit' ratio? From my figures I'd say this sits presently as either 7:48 or 7:24 (depending upon if you include term deposits with those basic 'get on with life, I just need to pay my electricity and buy some food' funds held in cheque and low-to-non interest accounts.

I keep coming back to what actually it is that we need to 'protect' - aka 'regulate'? I'd like that defined more clearly. I don't see any role for regulation (excepting fraud) in matters of commercial endeavour; let the risks fall where they will.

I do see a need for 'regulation as protection' to assist with our everyday transacting - i.e. that which must be done by means of money, which itself is something imposed upon us by government regulation.


Winton Bates said...

kvd: I agree that the proposed increase in shareholder equity is not 'flick of the wrist' stuff. Big changes need to be made to ensure that the owners of banks have sufficient 'skin in the game' to cover possible losses, rather than expecting taxpayers to pick up the tab.

Regarding numbers, I agree with your figure of shareholder equity of Australian banks of around 7% of total assets. I don't know my way around the banking statistics well enough to confirm your other numbers. The RBA Statistics on bank assets and liabilities seem to imply that in Feb 2013, total liabilities of banks (after adding shareholder equity) were accounted for as follows: shareholder equity, 6.6%; non-resident liabilities 16.8%; O/S operations liabilities, 5.0%; resident deposits, 58.2%; other liabilities to residents 13.3%. I have calculated equity as the difference between total assets and total liabilities. Such equity estimates are obviously subject to large error if there are errors in estimates of bank assets e.g. non-performing loans.

Regarding terminology, there is an important distinction between banks that are insolvent (i.e. unable to pay debts even after assets are sold) and those that have liquidity problems. A lot of the problems of the banks in the 19th century were due to liquidity problems i.e. 'runs'. That problem was more or less solved with introduction of central banks with lender of last resort powers and government guarantees, but that has encouraged banks to have lower levels of equity and thus to be at greater risk of becoming insolvent.

I think your concerns about every day transacting are parallel to my concerns to ensure that banks are sufficiently robust to be able to keep their promises to repay deposits according to agreed terms. The whole economic system breaks down when people/firms can't be trusted to keep their promises.

As regards regulation of commercial endeavour, it seems to me that firms that can call on taxpayers for funding if the value of their assets falls a few percentage points are more like creatures of government than commercial endeavours. Banking should be made to become a commercial endeavour.

It may be possible to make a case that as a condition for limited liability it should be illegal for any companies (not just banks) to make distributions to shareholders if equity falls below, say, 20% of the value of assets. Directors and shareholders don't seem to have strong enough incentives at present to use resources efficiently when firms are at risk of becoming insolvent.

Anonymous said...

Hello Winton

I thank you for your comments, and perhaps still we are talking terminology, but there's a couple of phrases you use, both in your comment and in the post itself, that I'd appreciate more information on.

First, in your post you refer to 'core banking functions of accepting deposits and making loans' which I totally agree with, but then you move away from this to discussion of banks as they now exist in all their gory.

Question for you: why not initiate a complete divorce of that core functionality from the rest of what our banks presently do?

I can see no public interest in supporting much of the present day functions of our banking system; just like BHP, they are prone occasionally to making horrendously bad investment decisions (bets?) in which the public has (nor should have) no real interest. But the differentiating factor is that they have also been granted a government monopoly over our savings and payments system - and it is their administration of this boring day-to-day stuff which I think matters to the public.

Second, your comment: "Banking should be made to become a commercial endeavour."

Question for you: If you define what is important to society that banking provides (aka your 'core functions'), then why do you add the further proviso that this be 'a commercial endeavour'? I would very much argue otherwise.

I think I'm politely getting to the point that your commentary is starting from the wrong end: you seem to be accepting present banking entities with their multitudinous, disparate roles and activities.

What I'm suggesting is that such of those activities as have a) been granted by government fiat, and b) are readily identifiable as being of public interest, be hived off from all the other quite legitimate activities that banks (as public trading companies) pursue.

Your '30% equity' would apply to all their activities - see Cyprus for how that worked out for depositors. Why not first get us to the point where we, the taxpayer, aren't backing those activities at all?

The reason I pursue this is that the ability to 'cut banks up' seems to me to be far easier from a legislation perspective than that required to get equity to 30% - which level itself would do absolutely nothing to protect against wayward commercial activities


Anonymous said...

Winton, sorry to go on about this.

Please just only read my last sentence to see what I'm getting at: for mine, be it 20-30-60% no change in equity in an of itself protects we the taxpayer from the vicissitudes of commercial activity.

It simply should not be regarded as a 'commercial activity' in the first place. 'Efficient' yes, but 'commercial' has no place in public policy, despite all the empty jargon of 'clients' and 'stakeholders' and 'mission statements'.


Winton Bates said...

Thanks for your comments, kvd.

My understanding is that attempts that have been made in the past to fence off the core functions of banking into a regulated sector have not been particularly successful. The problem is that the financial institutions that would be considered to be outside the fence can be just as important from a systemic viewpoint as those that are inside the fence. Some of the firms considered too big to fail in the GFC were not traditional banks. Their failure would still have impacted adversely on the core functions of the deposit-taking institutions.

Excessive bank regulation was one of the reasons why many people originally moved cash out of banks into money market funds.

Regarding your final point, I concede that we need governments to undertake a few core activities that cannot be conducted effectively through voluntary action of one kind or another. But we need to keep in mind that the only activities that governments can do better than ordinary people are those involving coercion - and there is no invisible hand of the market to ensure that they don't mis-use their coercive powers!

Anonymous said...

Winton, you introduce the concept of 'fence' - I expect that's the much-promoted 'ring fence'? That would be bound to fail, for the reasons you suggest.

We already sort-of have a ring fenced core banking operation; they are called building societies. I would strengthen the regulation of those entities, then place a government backing on their operation, and then remove any and all government support for what we now call 'banks'.

And I'd then remove all regulation of bank activities (excepting fraud, governance, etc.), because they would cease to have any significant public interest function.

You say that it was entities outside the banking sector which caused significant problems. This is true only if you close your eyes to the fact that they were intimately tied up in the balance sheets of the banks - either as subsidiaries in part or whole, or as willing investment vehicles for the banks, providing esoteric 'products' in which the banks were allowed to invest. I think they deserve each other, but that has absolutely nothing to do with your 'core banking functions'.

The banking operations which should be protected could easily be provided by Medibank, Australia Post, or our many small (and note that word!) building societies.

The rest of the major banking operations can basically be left to stew in its own juice as far as I care.


Winton Bates said...

kvd: I am now puzzled and sceptical.

I am puzzled that you have chosen the building societies as your core banking operation. My impression is that the building societies are more risky than the banks. They have historically tended to take on the low deposit loans that the banks will not touch and to pay higher interest rates in recognition of the greater risks involved in their activities. Perhaps my impression is out of date, or your proposed regulation would change matters.

I am sceptical because you seem to envision a situation where 'core banking' can be divorced from the rest of the economy. Even your heavily regulated core of the financial sector will be affected by changes in the world economy, with periods of high unemployment, declining property values etc. You seem to be assuming that we have a very wise government capable of regulating designated 'core banking' operations in great detail without causing them to lose business to market-driven financial intermediaries, and without requiring huge public subsidies or guarantees that would impose massive contingent liabilities on taxpayers.

In the real world, regulated banking usually involves governments in using regulated banks to channel funds in directions favoured by politicians.

Anonymous said...

Thank you for further comments, Winton. I accept your puzzlement, and skepticism - however I don't share either to the depth of your indicated concern.

Anecdote, I realise, but years ago (30 approximately) in Nowra my firm of Chartered Accountants (with me as responsible partner) established a local branch of the Bankstown Permanent Building Society. It accepted local money, lent on local property, was managed by a local (myself, who, chances were, was familiar both the the proposed property and the proposed borrower). Locals used it as a transactional facility, and were happy to see their money on-lent within the local community.

You could only borrow up to 75% of the property valuation (done by another local, the Shire Chief Building Inspector as it happened) or up to 85% with insurance. You could service this with repayments of 25% of your monthly income; something fairly readily established.

Aside: two of my three partners had family connections stretching back to the late 1800's, while I was a relative blow-in having arrived in the district at age 12. I mention this because, as always, trust is important in fiduciary relationships.

Anyway, Bankstown became Premier, became St. George, became Westpac - and the Lord only knows how little personal connection there is between each of the players now involved.

End of anecdote re 'core banking function' and building societies lending at the 'riskier end' of the market.

Nowadays I am uninterested as a taxpayer in guarenteeing ANZ's foray into SE Asia; nor the continuing saga of NAB trying to offload its UK operation for pennies in the pound; nor interested in Adelaide Bank being one of the heavy players in the futures market. I wish them well, but must add I'm particularly unimpressed with Bendigo Bank (for whose Nowra branch I am the largest single foundation subscriber) getting into bed with Adelaide.

You remain skeptical that such a simple thing as providing a transaction service backed by first mortgage residential property loans could be possible. I get the impression that you feel it better to re-darn the obvious holes in the socks worn by our 'Big Four' - which Four are now quite seriously 'too big to fail' despite (I assume) the best efforts of our betters in government and the Reserve Bank.

And if the solution is more regulation, then I'd suggest the 'problem' is not being properly analysed.

But I thank you for replying to my comments.


Winton Bates said...

I like your anecdote, but I'm not sure what it means. I hope your building society had a fairly high ratio of equity to total assets. I have the impression that the people who invested in building societies were once the legal owners, rather than depositors, but that situation seems to have changed over the years.

I am not sceptical about the possibility of providing a transaction service backed by first mortgage residential property loans. As far as I know that can still happen. I am just sceptical about government regulation, other than a requirement for relatively high equity.

I can also back up my claim of building societies being at the risky end with personal anecdotes. My wife and I bought our first home in Canberra 1n 1968 using building society funds with only a 10% deposit and a modest income to service the debt. I think that was fairly common at the time. A relative had a substantial amount invested in Pyramid building society when it folded - the family eventually got the money back (without interest, if I remember correctly). That was a case of a financial institution being small enough to be allowed to fail - and useful learning experience for a lot of investors!

I'm not sure where that leaves us. My focus is on systemic issues. I doubt whether anything the government might do to regulate building societies would protect people who invest in them if circumstances ever arose where one of the big four banks was allowed to fail.

Anonymous said...

'fairly high ratio of equity to total assets': that is not remotely how your garden variety building society worked. I remember the loan form contained an automatic $10 'application for shares'; can't recall if depositors had any such requirement, though probably so. On that basis we probably had 'equity' of less than 0.01%

Maybe you miscategorise current, unlent cash holdings as 'equity'?

Pyramid Building Society: basically failed because of inadequate, then ineffective (not to say inflammatory) government action, to combat a rumour started by a competitor. The business was sound, but the horses were frightened. It is instructive that your relative lost no money. Good thing it wasn't one of the 'new, improved' models our government condones such as Banksia.

Canberra: that would be the small country town which quietly financed (with taxpayer funds) the suburb-by-suburb removal of all asbestos roof insulation, a couple of years before it became a national health concern. My mother in law stayed with us while her entire suburb was 'degaussed'. I forget when it was ('85-'90?), but since that point I've assumed self-interest as the default Canberra position concerning the rest of us. (yes - I was a slow learner)

Canberra Building Societies: mostly lent (lend) to government employees; who in effect were (are) insulated from 'the real world' by virtue of their government employment. Back in the day, I would have probably lent them 110% based simply upon their lack of connection to (insulation from) the real economy, and their guarenteed employment.

It's all interesting. I think it just reflects our differing perspectives of what actually is 'the real world'.

As an example, it'll be very interesting to see how one of the last independent BS's - IMB (local to both of us) - plays out. Given history, I expect sooner or later they will be subject to false rumour and innuendo, and become then an easy 'rescue' target for whichever larger player sees them as a useful acquisition.

Subject of course to the usual thorough government oversight (cough), in the public interest cough).


Anonymous said...


I know, I know - off topic. But just to record that asbestos thing in case you think I made it up:

Canberra Times October 15 2011:

HUNDREDS of unsuspecting ACT businesses could contain the same sort of deadly ''fluffy'' asbestos removed from Canberra homes because the buildings were ''outside the scope'' of a $100million clean-up funded by the Federal Government in the late 1980s, an expert says.

The World Today September 14, 2012:

One of the primary recommendations of the review was to set up an independent agency and the Minister has announced that he's philosophically in favour of all the recommendations of this review and the first step within his department is to set up an Office of Asbestos Safety.

That thing will then have the job of doing a national audit because the point is, where is it? It's in all those old Telstra ducts, it's in your backyard. It's all over the place and that's a huge job.

Winton, all I can say is - you have to laugh. As opposed to taking up arms :)


Winton Bates said...

kvd: I was surprised that asbestos removal in Canberra was funded by the federal government. It must have happened just prior to self-government being imposed on the reluctant citizens of the capital city.

Self-government was intended to impose financial discipline. It might have succeeded in preventing federal public servants from paving the local roads with gold, but it didn't prevent local politicians from clocking up huge bills for local rate-payers. We were pleasantly surprised about the level of local government rates when we moved to the Shoalhaven -but don't tell Jo Gash!

Anonymous said...

Ah yes, Winton. Personally I love Canberra, and have had a long connection with it via my wife's family and now two of my children. I may even retire there to be closer to the grandbrood.

But it really is a different world. Many years ago on one of our visits to Tidbinbilla Nature Reserve I was using the public facilities just as two fellows were exiting - one loudly complaing to the other about there being no soap in the gents.

Re the asbestos, I wrote that from memory, before going to check if I was correct. Something I often do to keep the brain chugging.

And now I'll leave you in peace - but not before thanking you for your many thoughtful responses.