Growth-at-Risk

Growth-at-Risk (or GaR) links current financial conditions and economic conditions to the distribution of future economic outcomes. The downside of the distribution quantifies the magnitude of the expected losses in economic activity caused by deterioration in financial conditions, recognising the strong links the availability and price of financing for future investment, consumption and economic activity. GaR applies a forward looking distributional approach to outcomes, similar to a Value-at-Risk (VaR) measure as commonly applied to financial asset risk management.

The International Monetary Fund (IMF) estimates and regularly reports its estimates for the Growth-at-Risk for major international economies as part of annual Global Financial Stability Review. The IMF does not reguarly report estimates for Australia. The following provides the latest estimates for Australia's Growth at Risk based on the latest Australian financial conditions.

In order to highlight the marginal contribution to the growth distribution from current financial conditions, a baseline momentum measure is provided where the future growth distribution is a function the latest actual growth.


Historical Growth-at-Risk

The Growth-at-Risk distribution is represented in the following chart along with the actual real GDP growth. Generally the range captures the future eventual realised growth well, however unexpected events like the pandemic/COVID lock down and the subsequent rebounds exceed the prior distributions. The growth slow down even (although only temporarily) exceeded the growth implied by the forward looking financial market indicators.

P01-P99 provides the 99th percentile range of Growth-at-Risk months ahead. Similarly P10-P90 provides the 90th percentile range.

GaR and inflation cycles

Generally momentum (lagged GDP) implies a lower future risk distribution than financial market conditions. This is because fiinancial conditions are generally set to promote and support economic growth. It is only during periods of significant financial tightening that downside risks indicated by the Financial Conditions Indices have exceed momentum. There are four periods to highlight:

  1. The 1994 tightening
  2. The GST inflation spike (2000)
  3. The pre-GFC asset inflation bubble (2006)
  4. The post-COVID supply chain inflation spike (2024)

All periods coincide with unusually tight monetary policy implemented in response to persistent inflation outside the RBA target inflation band.

Methodology

The distribution of future economic activity is derived form of quantile regression, called quantile spacing. Two quantile regressions are compared in order to highlight the differences in the potential growth distributions from backward looking market momentum and forward looking financial market indicators.

1. Momentum Risk Model

where:

The model is fitted using quarterly data (where the FCI for the last month of each quarter) and is applied to the FCI for all months.

2: Financial Risk Model

where:

The three indices are derived as subsets of the Financial Conditions Index and include measures for Australia and the United States (full details of the constituents).

Each regression is estimates over the quantiles (q): 10th, 25th, 50th, 75th and 90th and over forecast horizons (h) of 6, 12, 18 and 24 months ahead.

Quantile spacing

While the general form of the quantile spacing estimation is consistent with the quantile regression, the difference is that while each quantile is estimated separately in quantile regression, the quantiles are estimated collectively in quantile spacing. Spacing ensures that the quantile estimates do not cross, that the estimates maintain rank order consistent with the quantiles they represent. Quantile spacing is estimated as:

Estimation results

Estimation results for all models are provided in the table below, not all explanatory variables are significant in the context of all quantiles, instead the influence of variables can differ in terms of their impact in strong and weak economic conditions.

References

Mr. Ananthakrishnan Prasad & Mr. Selim A Elekdag & Mr. Phakawa Jeasakul & Romain Lafarguette & Mr. Adrian Alter & Alan Xiaochen Feng & Changchun Wang, 2019, "Growth at Risk: Concept and Application in IMF Country Surveillance," "IMF Working Papers", International Monetary Fund, number 2019/036, Feb.

Luke Hartigan & Michelle Wright, 2021, "Financial Conditions and Downside Risk to Economic Activity in Australia," "RBA Research Discussion Papers", Reserve Bank of Australia, number rdp2021-03, Mar, DOI: 10.47688/rdp2021-03.