Macroprudential – Page 6 – Digital Finance Analytics (DFA) Blog

Managing Global Finance As A System

Andrew Haldane, Chief Economist of the Bank of England, gave the annual Maxwell Fry lecture on Global Finance at Birmingham University. There are some powerful observations here, relevant to the Australian context, as well as the potential to amplify risks associated with a more interconnected world. He also takes macroprudential discussions further.

Andrew’s main theme was the growing size and complexity of global capital flows between countries. He noted that ‘cross-border stocks of capital are almost certainly larger than at any time in human history’. And the apparent independence of domestic investment from domestic saving suggests that ‘measured levels of global capital market integration … remain at higher levels than at any point in history’. He discussed how this can be ‘double-edged’ from a financial stability perspective: it both shares risk (which can be stabilising) but also spreads and amplifies risk (which can be destabilising) – potentially generating ‘more frequent and/or larger dislocations’.

He argued that many lessons have been learned from the financial crisis, not least the need to ‘safeguard against systemic risk’. Yet when it comes to the fortunes of the international monetary system ‘it is far from clear that these lessons have been learned, much less that the international rules of the road have been reformed’. ‘Arguably, the rules of the road for this system have failed to keep pace with the growing scale and complexity of global financial flows’.

One of the consequences of the growth in cross border capital flows is ‘the steady rise in the degree of co-movement in asset prices over time’. Cross-border spillovers are becoming more important and global common factors more potent. A particular example is the behaviour of yield curves across countries. ‘To a first approximation, global yield curves appear these days to be dancing to a common tune’.

Andrew identified four areas where the global financial system could be strengthened:

a) Improve global financial surveillance, by tilting IMF surveillance away from monitoring individual country risk and towards multilateral surveillance and having more real-time tracking of the global flow of funds.

b) Improve country debt structures, for example by encouraging countries to issue GDP linked bonds, or Contingent Convertible (CoCo) bonds.

c) Enhance macro-prudential and capital flow management policies. For example, he suggests that ‘total credit follows a global cycle that has strengthened over time’ in which case ‘there may in future be a case for national macro-prudential policies leaning explicitly against these global factors’ taking international macro-prudential policy co-ordination ‘to the next level’. This next phase of macro-prudential policy may see measures ‘targeted at particular markets, as well as particular countries’.

d) Improve international liquidity assistance, for example by increasing the resources available to the IMF.

APRA Hoses Down Macroprudential

The Australian Prudential Regulation Authority (APRA) has published the opening statement given to the Senate Standing Committee on Economics by Wayne Byres today. He appears to deflect focus from macro prudential intervention.

Recent comments in the Reserve Bank’s Financial Stability Review about emerging imbalances within the housing market, and the need to reinforce sound lending practices, has been interpreted in many instances as Australia being on the verge of macroprudential interventions of the type that have been instituted in a number of other jurisdictions around the world, such as hard limits on certain types of loans, or minimum deposit requirements for borrowers. While we are very much still in the investigation stage, and have not yet decided what further action we might need to take, I would like to make two points in response to the general commentary currently taking place:

First, within our regulatory framework APRA generally seeks to avoid outright prohibitions on activities where possible: instead, our regulatory philosophy is to focus on institutions’ setting their own appetite for risk. We also use the regulatory capital framework to create incentives for prudent lending and ensure that, while institutions remain free to decide their lending parameters, those undertaking higher risk activities do so with commensurately higher capital requirements. That is not to say that we would never use the sorts of tools being employed elsewhere, but they are unlikely to be the first ones we reach for.

That brings me to my second point responding to potential risks in the housing market in this way is not new. We see it as standard supervision. In the period from 2002 to 2004, for example, there was a similarly strong run-up in house prices, and similar concerns about higher risk lending and emerging imbalances. We’re doing now what we did then: collecting additional information, counselling the more aggressive lenders, and seeking assurances from Boards of our lenders that they are actively monitoring lending standards. We’re about to finalise guidance on what we see as sound mortgage lending practice, and we’ve conducted a comprehensive stress test of the largest lenders. The sources of risk are different this time around – last time we were focussed on low doc and no doc lending – but the response of higher supervisory intensity and regulatory requirements in the face of higher risk activity is not new. It is APRA doing its job.

We have already highlighted the potential to adjust capital risk weightings as a mechanism to control risk, and it may be that the upcoming G20 will decide on lifting capital weights. However, we believe there is a role for targetted macroprudential measures in the investment housing sector as the argument is not just about risk per se, rather it is the fact that in the current low rate environment too much lending is going into unproductive establishing housing investment, rather than lending for productive growth. This point was well made in the speech yesterday by Philip Lowe, RBA Deputy Governor in an address to the Commonwealth Bank of Australia’s 7th Annual Australasian Fixed Income Conference in Sydney – Investing in a Low Interest Rate World. He concluded:

Very low global interest rates have been with us for some time. And it is likely that they will stay with us for some time yet.

Fundamentally, this reflects the low appetite for real investment relative to the appetite for saving.

These low rates are encouraging investors to buy existing assets as they seek alternatives to bank deposits earning very low or zero rates. Asset prices have increased in response.

Some of this is, of course, desirable and, indeed, intended. But the longer it runs on without a pickup in the appetite for real investment, the greater is the potential for new risks to develop. During this period, while we wait for the investment environment to improve, we need to be cognisant of potential risks of asset prices running too far ahead of real activity. This is true in Australia, as it is elsewhere around the world.

The underlying solution is for an improvement in the investment climate. Monetary policy can, and is, playing an important role here. But ultimately, monetary policy cannot drive the higher ongoing expected returns on capital that are required for sustained economic growth and for reasonable long-term returns to savers. It is instead government policy – including in some countries, increased spending on infrastructure – that has perhaps the more important role to play here.

RBA Still On The Low Rate Trip

The RBA minutes of the 7th October meeting are out. The themes are familiar, and they continue to signal an ongoing period of low interest rates, and the importance of lending standards.

Growth in the global economy was continuing at a moderate pace. Commodity prices, in particular iron ore prices, had declined over the past month. This was consistent with both the ongoing increase in iron ore supply and further weakening of the Chinese property market, which is an important source of demand for steel. Global financial conditions remained very accommodative and the Australian dollar had depreciated somewhat, largely reflecting a broad-based appreciation of the US dollar.

As expected, the domestic economy had grown moderately in the June quarter, following a strong March quarter result. The outcome was supported by strong growth in dwelling investment and steady consumption growth. Members noted that more timely indicators suggested that moderate growth overall had continued into the September quarter.

Faced with volatility in the labour force survey results, members based their assessment of the labour market on a range of indicators. These suggested that conditions in the labour market remained subdued but had stabilised somewhat this year. While forward-looking indicators pointed to modest employment growth in the months ahead, there was a degree of spare capacity in the labour market and it would probably be some time before the unemployment rate declined consistently. Wage growth was expected to remain relatively slow in the near term, which should help to maintain inflation consistent with the target even with lower levels of the exchange rate.

Members noted that the current setting of monetary policy was accommodative, with lending rates remaining very low and continuing to edge lower over recent months as competition to lend had increased. In this context, members discussed the importance of lenders maintaining strong lending standards and the ongoing dialogue between the Bank and APRA on the matter.

Continued accommodative monetary policy was expected to support demand and help growth to strengthen over time. To date, this had been most apparent in the housing market, where dwelling investment had picked up and was expected to remain strong following the rapid rise in housing prices and high levels of approvals. Credit growth had remained moderate overall, but in recent months there had been a further pick-up in lending to investors in housing. Despite the easing in financial conditions associated with the depreciation of the Australian dollar, the exchange rate remained high by historical standards – particularly given recent declines in key commodity prices – and was offering less assistance than would normally be expected in achieving balanced growth in the economy.

Given the information available, the Board’s judgement was that the current stance of monetary policy continued to be appropriate for fostering sustainable growth in demand and inflation outcomes consistent with the target over the period ahead. Members considered that the most prudent course was likely to be a period of stability in interest rates.

Looks like rates will remain on hold for a few months more yet, and macroprudential controls on investment lending appear likely.

Macroprudential, Revolutions and the RBA

Over fifty years ago, in 1962 Thomas S. Kuhn’s book The Structure of Scientific Revolution was published. It is an important work because if helps to explain how things work, and its findings I think are widely applicable beyond the scientific community.

Amazon says of the book “Kuhn challenged long-standing linear notions of scientific progress, arguing that transformative ideas don’t arise from the day-to-day, gradual process of experimentation and data accumulation but that the revolutions in science, those breakthrough moments that disrupt accepted thinking and offer unanticipated ideas, occur outside of “normal science,” as he called it. Though Kuhn was writing when physics ruled the sciences, his ideas on how scientific revolutions bring order to the anomalies that amass over time in research experiments are still instructive in our biotech age.”

His central thesis is that the evolution of ideas, where one set builds on the previous set does not adequately explain what happens in practice. Actually, new ideas often emerge away from the main stream, are often rejected by incumbents, thanks to positional power and authority, but some ideas, quite suddenly become the new normal, and become mainstream in their own right.

He argues that people in positions of power and influence tend to operate with a specific frame of reference, which makes it difficult for them to accept information which does not chime with their own views. Sometimes, though, revolutions do happen and as a result, we see quite sudden revolutionary changes in the accept norms.

I believe the RBA’s stance on macroprudential is an interesting example. How come that up to a couple of months ago, they were quite sanguine on the housing market, and dismissed macroprudential as a fad. Yet now, judging by recent comments, they are expressing concerns about the housing market, and we expect to see some form of macroprudential intervention before the end of the year. The data highlighting issues in the housing sector have been amassing for some time now, yet the RBA appears to have suddenly twigged and become a late convert.

Kuhn’s thesis seems to neatly explain the change.

Macroprudential Bites in the UK – Bank of England

The Bank of England (BoE) recently published their Q3 Credit Conditions Survey. This is the first edition since the recent Mortgage Market reforms were introduced, and the macroprudential controls were tabled. Looking specifically at secured lending to households they report that demand for credit has eased, and the proportion of higher loan to value loans has reduced. In other words macroprudential is biting!

After eight consecutive quarters of expansion, lenders reported that the availability of secured credit to households fell significantly in the three months to early-September.

The contraction in overall availability was reported to be driven by a changing appetite for risk and lenders’ expectations about house prices. Many lenders noted that operational issues associated with the implementation of the Mortgage Market Review had pushed down on credit availability over the summer. And some lenders commented that they had tightened availability a little in response to the recommendations made by the Financial Policy Committee to mitigate risks stemming from the housing market.

Credit availability was reported to have fallen in Q3 both for borrowers with loan to value (LTV) ratios below 75% and for borrowers with LTV ratios above 75%. Lenders also reported that they had become less willing to lend at LTV ratios above 90% for the first time since the question was introduced in 2013, and some noted that they had introduced policies which restrict lending at high loan to income (LTI) ratios.

Consistent with a tightening in credit availability, credit scoring criteria for granting household loan applications were reported to have tightened in Q3 and the proportion of household loan applications being approved fell.

Ireland Joins The Macroprudential Bandwagon

In the paper released by the Central Bank of Ireland, it is clear they have gone macroprudential! According to the Irish Times new rules will be applied to mortgage lenders in Ireland from 1 January 2015. A 2 month consultation period now starts, so they may get tweaked before implementation. The UK had announced parallel measures earlier.

New mortgage rules published today mean that most house buyers will have to have a 20 per cent deposit when applying for a home loan. The regulations come into force on January 1st.

The Central Bank is proposing that no more than 15 per cent of all new mortgages for private dwelling homes should have a loan to value (LTV) ratio above 20 per cent. This means that most first-time buyers are now going to be expected to have at least a 20 per cent deposit when buying a home.

In addition, it has also decided that just one-fifth of new mortgages should be issued above a level of three and a half times income (LTI).

In the case of buy-to-let properties, no more than 10 per cent of the value of all new loans should have an LTV above 80 per cent.

The Central Bank’s deputy governor said these measures should help to avoid another property crash in Ireland and dampen the rate of price rise currently being experienced in the market.

“Our research has shown there is strong evidence that mortgage losses are much higher where borrowers have a high LTV or LTI rate,” he said. “We believe that measures such as these are a standard part of a well regulated financial system and introducing these precautionary measures should contribute to a stable and well-functioning mortgage lending market.”

The regulator said the income caps would be “more binding” than the LTV ratios in a period of boom as pay levels could never keep pace with soaring property prices.

The LTV caps are not “completely counter-cyclical” as loan values will rise in line with property prices.

Certain exemptions are proposed to the new rules. These cover residual debt from home loans in negative equity, switcher mortgages, and home loans in arrears. Buy-to-let borrowers will also be exempt from the income restrictions.

Here is the Economist’s comparison chart, showing UK, Ireland and Australia – and we thought we had a problem!

IMF On Macroprudential – It Works!

In the just release IMF World Economic Outlook, as well as revising down growth estimates, they discuss macroprudential, highly relevant in the light of RBA comments. The main observations are:

there is evidence that macroprudential can assist in manage house price growth, and credit growth. Different settings should be applied to different types of purchases, e.g. differentiate first time buyers from multiple investors, but
it is less effective if the cause of extended price rises stems from overseas investors, who bypass local controls and credit policy, so specific separate measures may need to be used to target foreign investors
need to make sure business is not simply redirected to the non-bank sector, and
supply side issues also need to be addressed.

RBA please note! The comments in full from the IMF are below, and worth a read. In particular they cite a number of success stories, so macroprudential is perhaps more proven than many would like to admit.

Many countries—particularly those in the rebound group—have been actively using macroprudential tools to manage house price booms. The main macroprudential tools employed for this purpose are limits on loan-to-value ratios and debt-service to-income ratios and sectoral capital requirements. Such limits have long been in use in some economies, particularly in Asia.

For example, Hong Kong SAR has had a loan-to-value cap in place since the early 1990s and introduced a debt-service-to-income cap in 1994. In Korea, loan to-value limits were introduced in 2002, followed by debt-service-to-income limits in 2005. Recently, many other advanced and emerging market economies have followed the example of Hong Kong SAR and Korea. In some countries, such as Bulgaria, Malaysia, and Switzerland, higher risk weights or additional capital requirements have been imposed on mortgage loans with high loan-to-value ratios. Empirical studies thus far suggest that limits on loan-to-value and debt-service-to-income ratios have effectively cooled off both house price and credit growth in the short term.

Implementation of these tools has costs as well as benefits, so each needs to be designed carefully to target risky segments of mortgage loans and minimize unintended side effects. For instance, stricter loan-to value limits can be applied to differentiate speculators with multiple mortgage loans from first-time home buyers (as in, for example, Israel and Singapore) or to target regions or cities with exuberant house price appreciation (as in, for example, Korea). Regulators also should monitor whether credit operations move toward unregulated or loosely regulated entities and should expand the regulatory perimeter to address the leakages if necessary. For example, when sectoral macroprudential instruments are used to limit mortgage loans from domestic banks, they can be circumvented through a move to nonbanks (as in, for example, Korea) or foreign banks or branches (as in, for example, Bulgaria and Serbia). Macroprudential tools may also not be effective for targeting house price booms that are driven by increased demand from foreign cash inflows that bypass domestic credit intermediation. In such cases, other tools are needed. For instance, stamp duties have been imposed to cool down rising house prices in Hong Kong SAR and Singapore. Evidence shows that this measure has reduced house demand from foreigners, who were outside the loan-to-value and debt-service-to-income regulatory perimeters. In other instances, high house prices could reflect supply bottlenecks, which would need to be addressed through structural policies such as urban planning measures.

Macroprudential Tools could prove useful – RBA

In a speech in Melbourne, the RBA governor, Glenn Stevens said macroprudential tools could prove useful in helping to control the exuberant housing market. That said, he was still skeptical about their effectiveness.

He made the point that whilst monetary policy can’t solve every problem (i.e. interest rates alone) and there may be a need to take other steps if “at the margins they are helpful,”he didn’t consider macroprudential tools a simple solution to the problem, referred to in yesterdays Stability Review of strong investment lending. A reminder of the latest data, which we discussed recently.

He reiterated his concerns about the risks of investment loans, and highlighted the potential risks later, echoing yesterdays report.

No mention of macroprudential as a fad this time, which I guess is a step in the right direction. The IMF and OECD seem more convinced of the effectiveness of macroprudential. DFA’s view is we need them, and soon, alongside changes to negative gearing, and increased capital buffers.

It is interesting to note that U.S. regulators have announced that large banks will be required to hold more liquid capital to ensure they do not get into difficulty in a downturn. According to Reuters the eight biggest U.S. banks must boost capital levels by a total of $68 billion under these new rules. These rules are stricter than those under Basel III, and the banks have complained they will be put at a competitive disadvantage. They will need to hold tier one assets of 5%.

The Loan To Income Mess

Some time back we reported on the results of our household surveys, looking especially at the loan to income (LTI) data. This was prompted by the Bank of England’s move to limit banks abilities there to lend higher LTI loans. At the time we showed that at an aggregate level, LTI’s in Australia were higher than in the UK, yet despite this, there was no evidence of any local move to curb higher LTI borrowings, other than vague warnings from the regulators more recently. There is little relevant data published by the regulators on this important metric.

Today we delve into to the LTI data series in more detail. Interestingly the control of LTI’s was the preferred macroprudential tool of choice by BIS and others. The UK recommendation was to ensure that mortgage lenders do not extend more than 15% of their total number of new residential mortgages at Loan to Income ratios at or greater than 4.5 times.

In our surveys we ask about a households mortgage loan, and its total gross income. From this we can derive an LTI ratio.

So, to recap. this is the current picture of LTI, averaged across post codes for all household segments, and all states. There is a peak around 4.5 times, and a second peak above 6 times.

LTIAllStates Now, we can break the data out by state, and household segment, using the DFA survey data. The state specific data for NSW largely mirrors the national average.

LTINSW However, looking at TAS, we see some interesting variations. There the LTI’s are higher, reflecting lower incomes relative to somewhat lower house prices. We have adjusted the sample to take account of the smaller populations.

WA again shows variation, with a significant spread of higher LTI loans than the average.

LTIWA QLD shows greater concentration at lower LTI’s but then a second smaller peak at the upper end.

LTIQLD SA has quite a spike around 4.5 times, and a second peak around 6.5, again reflecting lower income levels in that state.

LTISA If we then start to look at segments, we find that affluent group, Exclusive Professionals, has a consistently lower LTI, compared with…

… Battlers …

… And the Young Growing Families. Many of the First Time Buyers are in this segment. Effective LTI’s above 7 times are significantly extended.

I won’t go through all the other segments now, but the analysis suggests to me that the LTI metric is significant, and a good leading indicator of risks in the system. Remember interest rates are at rock bottom at the moment, so households can keep their heads above water. But as we know rates may rise, and unforeseen events may change individual circumstances. Households with such high LTI’s have very little wriggle room. As the interim FSI report said “Since the Wallis Inquiry, the increase in housing debt and banks’ more concentrated exposure to mortgages mean that housing has become a significant source of systemic risk”. “Higher household indebtedness and the greater proportion of mortgages on bank balance sheets mean that an extreme event in the housing market would have significant implications for financial stability and economic growth”.