Bank Funding – Digital Finance Analytics (DFA) Blog

BBSW Funding Pressure Eases A Little More But …

Ironic that in the past two weeks a number of smaller lenders, including Macquarie and Bendigio/Adelaide Bank all hiked their mortgage rates, the interbank rate has eased. A couple of weeks back it was above 32 basis points now its down to 25 basis points.

So does this mean the risk of mortgage rate hikes have also abated from the majors?

Not necessarily as according to the AFR today, investors are moving funds from deposits to offshore locations. The share of bank funding from deposits is falling, thus they will need “an estimated additional $70 billion of funding as superannuation funds shift out of cash into international assets while indebted households draw down on their savings”.

This additional funding is required to support continued mortgage loan growth.

The household savings ratio is also falling thanks to the need to service mortgage repayments.

The AFR concludes “The banks faced two choices in response to slowing deposit growth – either lift deposit rates, or increase their use of the capital markets to meet demand for credit as their loan books have swelled to $2.6 trillion”.

“Either measure should result in higher cost of bank funding”

So still pressure to lift mortgage rates then.

ABC The Business Does Bank Re-Rating

The Business looked at the impact of S&P’s down grades on 23 smaller banks in Australia, and highlighted the impact on funding and competition, especially in the longer term. It will more than offset the bank levy the big banks will have to wear!

They also looked a funding costs and explained why mortgage rates may rise and the potential adverse impact on household debt.

Capital Flows to the Banking Sector

Deputy RBA Governor Guy Debelle, spoke at the Australian Financial Review Banking & Wealth Summit on “Recent Trends in Australian Capital Flows“. He highlights that Australian banks are still reliant on US funding sources, and this includes exposure to US commercial paper.

We therefore highlight that events there will impact banks here, especially changes in market rates (which are likely to be impacted by FED policy as we discussed earlier).

For more than a century, Australia’s high level of investment relative to saving has been supported by capital inflows from the rest of the world. These net capital inflows are the financial counterpart to Australia’s current account deficit. Foreign investment has been instrumental in expanding our domestic productive capacity and has been attracted by the favourable risk-adjusted returns on offer here.

Although net capital inflows have been a consistent feature of Australia’s balance of payments, the composition of both the inflows, as well as the outflows when Australians invest offshore, has varied substantially over time.

When I spoke about capital flows a few years ago, I discussed the significant changes in the composition of capital flows that had taken place since 2007. At that time, I highlighted three noteworthy developments: a marked increase in foreign direct investment in the mining sector associated with the mining investment boom; the significant change in flows to the Australian banking sector from sizeable inflows pre-crisis to around zero; and a substantial increase in foreigners’ purchases of Australian government debt.

To a large degree, these trends have continued over the past three years. But under the surface, there have been some significant changes in the composition of these flows in recent years.

The aggregate pattern of capital flows to the banking sector has not changed materially since I last spoke on this topic. Since 2014 – and indeed over the period since the financial crisis – there have been minimal net capital flows to or from the banking sector. Following the shift away from offshore wholesale debt towards domestic deposits that took place in the wake of the global financial crisis, the funding composition of banks has remained relatively stable. But notwithstanding this stability, recently there have been two noteworthy developments relating to short-term debt, both stemming from regulatory reforms.

Firstly, over the past year or so, Australian banks have reduced their short-term debt issuance in preparation for the introduction of the Net Stable Funding Ratio (NSFR) next year. The NSFR provides an incentive for banks to shift to sources of funding considered to be more stable and away from sources such as short-term wholesale liabilities.

Secondly, the composition of Australian banks’ short-term offshore funding has also changed following the implementation of US Money Market Fund (MMF) reforms by the Securities and Exchange Commission in October 2016. As a result of these reforms, the value of assets under management of prime MMFs (those that lend to banks) has fallen by US$1 trillion or around 70 per cent over the past couple of years. Some prime funds have switched to become government-only funds, that is, funds that invest only in US government debt. At the same time, investors have allocated away from prime funds to government-only funds. Although prime MMFs have maintained their exposure to Australian banks relative to banks globally (at around 8 per cent of total MMF exposures to banks), their holdings of Australian bank debt have declined from around US$100 billion to under US$30 billion currently.

However, in aggregate, Australian banks have continued to raise almost as much short-term funding from US commercial paper markets, despite the decline in MMFs. They have been able to tap other investors, in particular US corporates with large cash holdings, such as those in the technology sector.

Bank Spreads Have Improved, Thanks To More Expensive Home Loans

In the RBA Bulletin, released today, there is a section which shows major bank margins have improved. Essentially, the story is one of falling deposit rates, plus change in mix, and rises in home lending rates independent of the cash rate in recent months. Consumers are bearing the burden whilst big business lending rates and margins are being compressed thanks to competition from overseas banks. The big banks are benefiting the most from the margin improvement, which shows again their market power.

Consistent with the large fall in interest rates on term deposits, the level of term deposits in the system declined in 2015, mostly due to some maturing deposits not being rolled over. In contrast, transaction and at-call savings deposits grew strongly in the year .

Throughout 2015, banks also adopted pricing strategies aimed at reducing deposits from institutional depositors (such as superannuation funds), which are more costly to banks under the LCR framework. The change in the mix of deposit funding lowered the cost of those funds by 3 basis points owing to particularly strong growth in transaction deposits, which carry lower interest payments. In part, this reflects the rapid growth of mortgage offset account balances through 2015, where funds are typically deposited in zero-interest rate accounts but are used to reduce the calculated interest on the associated mortgage. One implication of the increased use of such accounts is the high ‘implied’ cost of funds for banks – equivalent to interest forgone on mortgages. Interest rates on mortgages are much higher than those on deposit products, so banks implicitly pay their customers the mortgage rate on funds held in offset accounts. However, money held in offset only accounts for about 6½ per cent of at-call deposits.

They make the point that housing rates generally declined in line with the cash rate in the first half of 2015, with the average outstanding interest rate on mortgages falling by around 50 basis points over that period. In the second half of the year, however, banks adjusted their lending rates such that the average outstanding housing interest rate for investor loans was only modestly lower over 2015, while rates for owner-occupiers declined by roughly 30 basis points over the year. Interest rates on investor loans were increased midyear, following concerns raised by APRA about the pace of growth in lending to investors. Increases in investor lending rates ranged from around 20–40 basis points, and were applied to both new and existing investor loans.

In November, the major banks raised mortgage rates across both investor and owner-occupier loans by 15–20 basis points, citing the cost of raising additional equity to meet incoming regulatory requirements. Of particular relevance, the Financial System Inquiry’s Final Report recommended higher capital requirements for banks using ‘advanced’ risk modelling (the major banks and Macquarie Bank) in order to reduce a competitive disadvantage relative to other mortgage lenders (FSI 2014). The other Australian banks similarly increased mortgage lending rates, despite not facing the same regulatory costs as the major banks.

Business rates generally fell by more than the cash rate in 2015, with large business rates falling by around 70 basis points and small business rates by around 60 basis points. These lending rates remain at historic lows. Banks reported that declines in business rates beyond the changes in the cash rate were driven by intense competition for lending, including from the Australian operations of foreign lenders.

The major banks’ implied spread, being the difference between average lending rates and debt funding costs, increased by around 20 basis points over 2015. This change was driven in roughly equal parts by the decline in average funding costs relative to the cash rate, and an increase in the average lending rate. However, lending rates and debt funding costs tend to move in line with each other in the longer run. The contribution to the aggregate implied spread from higher lending rates was entirely due to increases in housing lending rates, with the implied spread on housing lending now higher than the previous peak in 2009. However, the measure of funding costs used to calculate implied spreads does not account for the increased share of relatively expensive equity funding. As such, the increase in the implied spread for housing lending is likely to overstate the true change in major banks’ margins for this activity. Implied spreads on business lending declined over 2015. Consistent with strong competition, implied spreads on large business lending have returned to pre-global financial crisis levels, when there was strong competition, business conditions were highly favourable and risk premia were compressed. Much of the competition is coming from foreign banks, with the average rate on business loans written by foreign banks significantly lower than the rate being charged by Australian banks.

The average implied spread on other Australian banks’ lending has been around 25 basis points lower than for the major banks since 2005. However, there is considerable variation in implied spreads of the other Australian banks, driven more by the high variation in lending interest rates across banks than variations in funding costs. In contrast to the major banks, the spread on other Australian banks’ lending for housing declined over 2015 with their lending rates falling by more than their funding costs. The other Australian banks’ spread on business lending also decreased in 2015. The spread on business lending remains higher for the other Australian banks, which reflects the fact that these banks generally lend more to smaller business than the major banks, and do not compete as heavily with the major and foreign banks on large business lending.

External Forces May Impact Bank Funding – And Households

On the international horizon there are two potential events which may impact Australian bank funding. Today we consider the potential impacts on the banks, and households. First, it is likely the FED will start to lift interest rates later in the year as they adjust towards more normal rates. Second within a couple of weeks the Greek situation will crystalise, with either a negotiated debt settlement or an early exit. Both these (not totally unconnected) issues could play on bank funding because the large banks here are still quite reliant on accessing the international financial markets.

In recent times funding costs have fallen from their very high levels during the GFC. Australian bond spread mirrors the global picture.

The question is what happens if the financial markets react negatively to the news from the USA or Europe? At very least we can expect greater volatility, and the risk premium on funding costs would likely be raised. This would potentially translate to higher funding costs for the banks because they do rely on the global financial markets (we have been a net importer of funds to support the banks for years).

However, the latest data shows that the banks have a greater proportion of funding from deposits compared with pre-GFC, and there has been a corresponding fall in short term debt funding.

So we think the Australian Banks are quite well placed, despite the potential need to raise an additional $20-30bn to meet expected changes to their capital ratios (work in progress but it is certain to rise). They have already shown their willingness to drop deposit rates more than the market, and we think it is likely this would go further.

In addition, they could reduce the discounts on mortgage lending, as currently they are quite high.

They could also throttle back on their lending – especially for housing – to better match deposit growth to lending growth. Finally, they have access to the RBA “emergency” fund if needed – The Committed Liquidity Facility (CLF).

The Reserve Bank is providing a Committed Liquidity Facility (CLF) as part of Australia’s implementation of the Basel III liquidity standards from 1 January 2015. Consistent with the standards, certain authorised deposit-taking institutions (ADIs) are required by APRA to maintain a liquidity coverage ratio (LCR) at or above 100 per cent. These ADIs may seek approval from APRA to meet part of their Australian dollar liquidity requirements through a CLF with the Reserve Bank. In consideration of the Reserve Bank’s CLF commitment to an ADI, the ADI must pay a monthly CLF Fee in advance to the Reserve Bank.

Industry insiders estimate the capacity of the facility could be as high as $300 billion, a substantial amount. In effect the banks are backed by a Government guarantee.

So, we think that the wash through of these international issues will not create major financial stability problems locally, but there may well be higher costs to savers and borrowers, and potentially Australian tax payers, if the RBA is called on to assist with liquidity, or worst case the deposit insurance scheme is called upon if a bank were to get into difficulty. Currently deposits up to $250,000 in ADI’s supervised by APRA are covered.

Two other points to highlight. First, household savings ratio are on their way down, from their highs post GFC. We would expect households to hunker down and save more if there was an external shock, thus bolstering bank deposits (and a likely flight to quality, as we saw in the GFC). We do not think a run on an Australian bank is likely.

Second, bank net margins are compressing thanks to the severe competition in mortgage lending. This dynamic may change if there was an external shock, as demand for mortgages eased, providing some relief.

In addition the US Australian dollar exchange rate would probably drop, providing some relief. However, second order impacts, namely slowing economic activity, falls in confidence, and rising unemployment which may follow from a global shock, in turn have the potential to impact the banks much more, because it would translate into risks in the housing sector basket, where they have been placing their eggs in recent times.

Bank Spreads Have Increased By 35 Basis Points

Continuing our analysis of bank margins, we have updated our industry model, with the latest funding and product data. At an aggregate industry level, we see that the average home lending rate has remained static (because whilst there are substantial discounts for new loans, the bulk of the back book has not seen any rate reduction) since 2013. The RBA last reduced their cash rate in August 2013, and the benchmark rate has remained static since then.

However, savers have see their returns falling thanks to deposit repricing initiatives. Between September 2013 and now, the average deposit return has dropped by 35 basis points. As we explained, banks are less reliant on deposits, and can get cheaper funding from other sources (and the recently announced QE in Europe will make funding even cheaper).

So, despite the fact that the banks are unlikely to be able to reduce their provisions much further (as they did last year) to bolster profits, and their increased capital requirements, the highlighted net increase in margins bodes well for bank performance, though at the expense of borrowers, who are not enjoying rate reductions, and depositors who are seeing their interest rates continuing to fall.

New Liquity Rules Will Reduce Deposit Rates Further

RBA Assistant Governor (Financial Markets) Guy Debelle speaking at the 27th Australasian Finance and Banking Conference in Sydney on 16 December 2014 summarised the changes to bank liquidity which are translating to lower deposit rates for savers in 2015. We already highlighted the falling deposit rates for savers (despite no change in the RBA rate).

“Today I will talk about the imminent arrival of the revised liquidity regime for the Australian financial sector. I will recap some of its features, particularly how they relate to the Reserve Bank, and discuss some of the impact that it is having on market pricing.

An important aspect of the Basel III liquidity standard, the Liquidity Coverage Ratio (LCR), comes into effect in under one month’s time at the beginning of 2015. The LCR requires that banks hold sufficient ‘high quality liquid assets’ (HQLA) to withstand a 30-day period of stress. The amount of HQLA a bank needs to hold is determined by the composition and maturity structure of its balance sheet. The more liabilities that run off within that 30-day window, the more HQLA that needs to be held. At the same time, particular types of investors or depositors are assumed to be less stable than others (in terms of their likelihood of withdrawing funds), which also results in a greater need for liquid assets.

As has been known for some time, the Australian financial system does not have an especially large stock of HQLA. The only instruments that have been deemed to meet the Basel standard of liquidity are debt issued by the Commonwealth and state governments (CGS and semis) along with cash balances at the Reserve Bank. The banking system’s overall liquidity needs are greatly in excess of what could reasonably be held in those assets. To put some numbers on this, APRA has determined that for next year, the Australian banking system’s liquidity needs amount to $450 billion. The total stock of CGS and semis on issue currently amounts to around $600 billion. If the banks were to attempt to meet their liquidity needs solely by holding only CGS and semis, a number of problems would arise. Firstly, any attempt would likely be in vain, because there are a large number of other entities which are required to or want to hold CGS and semis too. Second, in the process of trying to do this, the liquidity of the market for these securities would be seriously compromised. This would be completely self-defeating as the overall aim is to have the banks hold more liquid assets.

To address these circumstances, an important component of the liquidity regime in Australia is the committed liquidity facility (CLF) where, on the payment of a 15 basis point fee, banks will be able to obtain a commitment from the Reserve Bank to provide liquidity against a broad range of assets under repurchase agreement.

APRA has recently determined that the total CLF requirements of the Australian banking system for 2015 amount to around $275 billion. This amount was determined by first assessing that the amount of CGS and semis that could reasonably be held by banks without unduly affecting market functioning was $175 billion. The Reserve Bank provided this assessment to APRA. The CLF amount is then simply the difference between this and the overall liquidity needs of the system.

The banks that require a CLF from the Reserve Bank sign a deed of agreement with us and pay their fee before the end of this year. Then from the beginning of next year, the arrangement comes into effect.

I have talked before about some of the impact on pricing in various markets of the new liquidity regime.[⁴] We have attempted to limit the impact on the price of CGS and semis, but necessarily, because the banks are holding more of these securities than previously (Graph 1), the price is higher (and the yield lower) than would otherwise be the case.

Graph 1

Overall, the impact of the LCR on market pricing is relatively small. The larger changes have been around deposit pricing and the terms and conditions of deposits, which I will come to shortly, but there have been some other effects which are worth commenting on.

Firstly, a less discussed aspect of the liquidity standard is the requirement for a demonstrated internal liquidity transfer pricing model for banks. This has required banks to fully reflect the liquidity cost in the price of the various services they offer customers. This has resulted in a change in the price and/or terms and conditions of a number of facilities. One noteworthy example is a line of credit where, in the past, banks often did not factor into the price they charged for this facility, the potential draw on liquidity this entailed, particularly in a stressed situation. On the other hand, longer fixed-term deposits are more attractive to banks and consequently have been repriced upwards (see below).

A second impact which has been evident more recently is a widening in the spread between bank bills and OIS (Graph 2). In the depths of the crisis, such a widening was often an indicator of stress in the financial situation. But that does not appear to be the case currently as other indicators of bank creditworthiness are little changed, including spreads on longer term borrowing and CDS premia.

Graph 2

Instead, our assessment is that in large part, this reflects the new liquidity regime combined with some other dynamics in the market. The graph shows that the widening has been most pronounced at the longer bank bill maturities, and indeed is quite small for a one month bank bill. This is because issuing a one month bill has little attraction to a bank: its liquidity cost is relatively high as its maturity is likely to occur within the 30-day liquidity window. Hence a bank would need to hold HQLA of similar size to the amount of funding the bank bill raised. Instead, there is a greater incentive to issue at longer maturities and so the spread on 6-month bills has widened by more as there has been greater supply of such paper.

Over the past two months, the original term to maturity of bank bills and certificate of deposits on issue has changed noticeably, with the stock of 6-month bills increasing by $7.3 billion (11 per cent) and 12-month bills by $1.4 billion (43 per cent). In contrast, the stock of outstanding bills with an original tenor shorter than five months has declined by a total of $9.7 billion (8½ per cent).

At the same time, the cost of Australian dollars in the forward FX market has been quite elevated. This high price in the forwards market has been due to a number of factors including the tendency of hedge funds to fund their Australian dollar shorts in this market, as well as an increase in the use of this market by foreign bank branches to fund Australian dollar lending.

Historically, Australian banks have tended to raise a significant share of their short-term funding in foreign markets, mostly in US dollars, and then swap them back into Australian dollars to fund their Australian dollar-denominated asset base. They would swap these foreign currency funds when the cost was sufficiently attractive, leaving it in foreign currency in the interim. Under the new liquidity regime, the cost of short-term foreign currency funding is higher, so this is less attractive. Combined with a higher swap cost, the all-in cost of short-term offshore funding is higher and hence domestic issuance is relatively more attractive. As a result we have seen more of it, which has contributed to the widening in the spread to OIS.

Finally, I will return to the impact of the LCR on deposit pricing. Graph 3 shows the evolution of the funding mix of Australian banks over the past decade. The rise in the share of deposit funding from 2008 is readily apparent, as is the decline in the share of short-term and long-term wholesale funding. The growth in deposits is now of a similar pace to that in bank lending, having been considerably faster over recent years. As a result, the deposit share of funding has levelled off.

Graph 3

The increase in deposit funding was in part a result of the increased returns on offer, as banks actively sought this outcome by offering higher interest rates. (It also reflected a shift on the part of investors for the perceived safety of a bank deposit.) The interest rate on both at-call and term deposits rose markedly compared with money market rates of equivalent maturity (Graph 4). As you can see from the graph, this process of paying higher deposit rates has largely run its course.

Graph 4

Within this overall repricing, there have been some changes in the mix of deposit rates and products as a result of the introduction of the LCR. As I mentioned earlier, banks have an incentive to reduce the amount of liabilities that roll off in less than 30 days. Deposits which are deemed to be subject to high run-off rates and those which are callable within 30 days will be more expensive for banks. Banks are therefore working towards converting many of these less stable deposits into a more stable deposit base. For example, retail and SME deposits are deemed to be ‘stickier’ than institutional deposits. Part of this transition is being induced by price signals: interest rates offered on new or existing deposit products which are deemed to be more stable are rising relative to interest rates on products deemed to be less stable.

These types of changes appear to have accelerated recently as we draw closer to the implementation of the LCR and probably still have some way to run. To date, we have seen only a few banks offer notice of withdrawal accounts to customers. These accounts require the depositor to provide the bank with 31 days or more notice of a withdrawal (obviously 31 days is one day longer than the 30-day liquidity stress period). Interest rates offered on these accounts are among the higher rates offered in the deposit market. It may be that we see a broader move to these types of accounts or changes in terms and conditions on existing accounts through the course of next year.

We have also seen a fall in the growth of term deposits relative to transaction and at-call deposits over the past few years. In fact term deposits as a share of banks’ funding has been falling while transaction and at-call deposits have been growing strongly. Part of this is because a flattening of the yield curve has made investors less inclined to invest in longer term deposits. But in part it is because under the LCR, some transactional and operational deposits are subject to lower run-off rates than deposits that are largely attracted by higher interest rates. Indeed, there has been some indication that banks have been transitioning depositors into deposit products treated more favourably under the LCR.

But banks are not limited to just changing their deposit offerings. We could see them look for more opportunities to package retail deposits with other products as the deposits of customers that also have other relationships with the bank are deemed to be more stable under the LCR.

So to conclude, the full implementation of the new liquidity regime in Australia is imminent. From the beginning of next year, banks in Australia will be fully subject to the Liquidity Coverage Ratio. This has already had an impact on the pricing and nature of a number of financial products, as well as the structure of bank liabilities. While the bulk of the impact may be behind us, there are still a number of changes in the pipeline, particularly around deposits.”