Geof Mortlock argues Minister of Finance Nicola Willis should commission an independent review of the RBNZ's DTI policy coordinated by Treasury

By Geof Mortlock*

Recently, the Reserve Bank (RBNZ) released its proposal for restrictions on bank lending by reference to borrower debt-to-income (DTI) level. Under this proposal, banks will be restricted in a proportion of their lending based on a borrower's DTI as well as a borrower’s loan-to-valuation ratio (LVR).

The RBNZ has argued that these restrictions are necessary to maintain the stability of the financial system by reducing the probability of bank losses attributed to borrower default on residential property loans and excessive house price volatility. However, I believe the proposed lending restrictions are not justifiable by financial stability concerns and that, if financial stability is truly the objective, there are more efficient means of meeting that objective than the DTI restriction.

Rather, it is my view that the RBNZ has proposed the DTI restriction in an attempt to moderate any future increase in residential property prices as a goal in its own right and to address concerns about borrower debt stress, regardless of whether this is necessary to maintain financial system stability. This is an example of the RBNZ's overreach on its macroprudential policy function.

Before discussing the DTI proposal, it is useful to remind ourselves about what macroprudential policy is supposed to be about. Drawing on International Monetary Fund (IMF) and Financial Stability Board (FSB) guidance, macroprudential policy is defined as the use of prudential regulation to limit financial system risk. Systemic risk is the risk of widespread disruption to the provision of financial services that is caused by an impairment of the financial system, with negative consequences for the real economy.

Systemic risk is generally recognised as having two key elements: vulnerabilities related to the build-up of risks over time and vulnerabilities from financial institution interconnectedness.

Internationally, macroprudential policy seeks to achieve financial system stability through three main intermediate objectives: (a) increase the resilience of the financial system to economic shocks through countercyclical capital buffers; (b) reduce the build-up of systemic vulnerabilities; and (c) control vulnerabilities within the financial system that arise through interlinkages and common exposures.

As with any policies, macroprudential policies should be assessed on a cost-benefit basis to ensure efficacy and to avoid unintended consequences and costs. The least-cost path to achieve financial stability objectives is the best approach.

In my assessment, the RBNZ is using its macroprudential policy function for purposes that go beyond financial stability concerns and that, in the process, it is imposing regulatory inefficiencies on the financial system and economy. The current proposal is ostensibly being justified on financial stability grounds, yet the rationale for the financial stability concern is unproven.

Stress tests of banks conducted by the RBNZ have repeatedly shown that banks are resilient to even severe house price shocks and sharp increases in the rate of unemployment.

The RBNZ has not demonstrated the case for using DTI or LVR restrictions on the grounds of financial stability.

I suspect that the DTI initiative is not really motivated by financial stability concerns per se, despite the narrative put forward by the RBNZ. Rather, I believe the DTI/LVR policies reflect the RBNZ’s desire to moderate house price inflation for broader ‘political’ purposes, such as concerns relating to housing affordability and excessive household indebtedness.

There are indeed legitimate concerns over the excessive levels of residential property prices. Rising household debt is also a major concern. However, these are not matters that fall within the purview of the RBNZ. Its only mandates are to keep CPI inflation within the target range (which it has failed so badly to do due in part to mismanaged monetary policy) and to maintain the stability of the financial system.

The most appropriate policies to address residential property price concerns lie outside of the RBNZ’s mandate. For example, appropriate policies are likely to focus on: (a) increasing the allocation of urban land zoned for residential construction; (b) strengthening the capacity of the building sector to build more properties; (c) relaxing unnecessary building code and consent requirements; and (d) reducing the level of immigration. None of these policy issues lie within the remit of the RBNZ.

The RBNZ DTI proposal is an example of overly prescriptive and poorly costed regulation. It will create inefficient distortions to bank lending and will prevent many people who could adequately service a loan from accessing housing finance due to arbitrary cut-off limits. It disregards the many factors which a bank will take into account when assessing the capacity of a borrower’s credit worthiness.

This type of regulation is reminiscent of ‘Muldoonism’ and the pre-1989 approach to bank regulation - blunt, cumbersome, and poorly designed - with all the adverse impacts and economic and financial distortions to which such an approach can give rise.

Micro-management of bank lending decisions should have no place in a well-functioning market economy and financial system. Indeed, given that the RBNZ demonstrably lacks people with banking expertise and experience (including, remarkably, its senior management team), it is rather disturbing that they see themselves as being better qualified than bankers to assess the credit worthiness of borrowers.

If there is a legitimate concern over financial stability associated with potential housing price bubbles and high levels of household debt, the cost-effective and efficient answer does not lie in prescribing arbitrary limits on lending by reference to DTI and LVR ratios.

Rather, it lies in ensuring that banks have sound governance, robust risk management systems, and sufficiently high capital buffers to absorb losses. Despite the fact that severe stress tests do not identify vulnerability in the banks, the RBNZ has already imposed on banks very high capital ratio requirements by international standards.

If (which I doubt) further initiatives are needed to protect the financial system from housing price disturbances, then the more cost-effective option is likely to include reassessing the calibration of risk weights in bank capital ratios.

For example, risk weights could be fine-tuned to more accurately reflect potential losses based on DTI and LVR factors. This would be much less distortionary, less arbitrary, and involve lower moral hazard risk than the RBNZ’s cumbersome proposal for quantitative DTI lending restrictions.

I am concerned that the RBNZ has not demonstrated sufficient analysis of the financial stability justification for this proposal and has not undertaken a sufficiently rigorous cost-benefit analysis of its impact. It is also concerning that the RBNZ’s regulatory proposals are not subject to robust independent assessment. In that respect, it is my view that The Treasury does not provide sufficient scrutiny of RBNZ (or other) regulatory proposals.

Unlike in Australia, for example, where regulatory proposals are subject to robust independent cost-benefit assessment by a well-resourced government agency dedicated to that function, New Zealand lacks adequate cost-benefit assessment frameworks.

Far too often, the RBNZ and other government agencies develop cost-benefit assessments and regulatory impact statements only once they have already made up their mind what they want to do. They 'reverse engineer' the assessment to justify their preferred position. The RBNZ cost-benefit assessments are typically light on analysis, too late in the regulatory process to be truly useful, and largely self-serving.

Treasury, as the agency responsible for exercising scrutiny over regulatory impact statements, does an inadequate job. It is under-resourced for the task, lacks the expertise (and makes little effort to engage external expertise to assist in the task), and is not sufficiently assertive in pushing back on regulatory proposals. Ministers are also too often missing in action in such matters.

I believe the DTI policy needs to be fundamentally reassessed. The Minister of Finance should commission an independent review, coordinated by The Treasury, to evaluate the efficacy of the proposal and to assess whether: (a) it is justified on financial stability grounds; (b) it might create unnecessary distortions to lending and flow-on effects to the real economy; (c) it might unnecessarily and arbitrarily impede borrowers from accessing finance even when they can meet a bank’s credit requirements; and (d) whether alternative options (such as recalibrating bank capital ratio risk weights) would provide more cost-effective and less distortionary means of meeting financial stability objectives.

I hope that Nicola Willis and David Seymour will pay close attention to these considerations.

The DTI proposal raises a bigger issue about the performance of the RBNZ generally and the adequacy of its governance, management, transparency, and accountability. But that is a topic for another article.

*Geof Mortlock is a consultant on economic and financial policy, drawing on many years of experience as former senior staffer at the RBNZ and APRA. He undertakes regular consulting assignments globally for the IMF, World Bank, KPMG and other international organisations.