‘Normalisation,’ if you can keep it: the BoE on the QT
Plus, Australia’s 2021 US equities buying bender
I was struck by this headline in the FT, which is more reminiscent of the those you would expect to see late-, not mid-cycle:
From the accompanying story:
If you think it is slightly odd for a central bank to be tightening at the same time as warning the economy is set to stall, you would be right. It is not often central banks signal a downturn, much less tighten into one.
The BoE has already returned its policy rate to above pre-pandemic levels, having commenced raising interest rates back in December and quantitative tightening back in February. Like the RBA, the BoE plans to let gilt maturities contract its balance sheet, although has also set a threshold for its policy rate of 1.00% at which point they would consider a program of active selling, which is only 25 basis points away. You can see a member of the BoE MPC make their case for tightening here:
Having front-run the global tightening cycle, the BoE will be worth watching, both for the effects of active QT, if it comes to that, but also for a prospective return to a new global easing cycle. David Smith suggests that the BoE is being strategic and trying to hose down inflation and wage expectations through its forecasts, but this seems a little too clever too me.
Next week sees the release of Australia’s Q1 wage price index, which our model has at 0.8% q/q and 2.5% y/y compared to 2.3% y/y for the final quarter last year. At his press conference following the May Board meeting, Governor Lowe gave a critical to role to wages growth in intermediating between transitory inflation pressures attributable to temporary supply shocks and ongoing inflation pressures embedded in the wage and price setting process. Given the inflation outlook, the RBA’s former enthusiasm for wages growth at or above 3% can be expected to fade somewhat. Whereas 3% annual wages growth was formerly seen as a threshold for tightening, this threshold condition always had the dynamic relationship between inflation and wages backwards. The RBA’s May tightening effectively concedes that point. Wages growth will still be an important determinant of the RBA’s tightening cycle, but not in the way the RBA formerly led us to expect.
Sam Wylie suggests that the failure to wait on wages growth means that the ‘RBA’s reputational capital has been spent like water.’ Terry McCrann has also been on the warpath, saying ‘The Reserve Bank’s latest forecasts for the economy are a humiliating admission of failure.’ But as I argued last week, if our benchmark is macroeconomic outcomes, RBA policy is more appropriately viewed as too successful rather than a failure. While I have significant issues with how we got here, the earlier than forecast commencement of a tightening cycle is not in itself a policy failure. If the RBA’s earlier forward guidance had been correct and it did not tighten policy before 2024, what would that have implied about macroeconomic conditions, both currently and prospectively? Until recently, my own view had been that policy had not been aggressive enough given their forecasts. But I still feel comfortable having erred on the side of over-doing it. Given where we were six months ago, we have landed in a good place.
Matt Cowgill’s visualisation of the change in the RBA’s forecasts between the February and May Statements on Monetary Policy, apart from showing the biggest ever forecast revision to the inflation outlook, is remarkable for showing an across-the-board downgrade in near-term expectations for the real side of the economy. This is indicative of an economy being hit by inflationary supply shocks rather than a positive demand shock:
The US CPI rose by 0.3% in April compared to 1.2% in March. This lets us update our Cleveland Fed model of the Australian trimmed mean inflation rate for Q2. The Cleveland Fed measure fell relative to its average in Q1. This would leave the Q2 trimmed mean rate in Australia at 1.3% q/q and 4.3% y/y versus the RBA’s forecast of 4.5% y/y, but we still have to see how the US trimmed mean evolves over May and June before concluding that the local trimmed mean rate will come in below the RBA’s forecast.
This week, I attended a CIS seminar where Chris Murphy presented his modelling of Australia’s COVID-era fiscal stimulus. The key finding of the paper is that these fiscal measures over-compensated households and business for pandemic-related income losses by a factor of around two to one and created perverse incentives for a representative firm to restrict output. Compared to ‘no-pandemic’ and ‘no-discretionary stimulus/automatic stabilisers only’ counterfactual scenarios, the fiscal stimulus largely accounts for the current spike in inflation. Indeed, Murphy’s baseline scenario called the timing and the magnitude of the lift-off in both inflation and interest rates well before the fact (the modelling results were first presented in late October last year).
Putting that predictive success to one side, a significant limitation of the model is that it largely abstracts from what was happening with monetary policy. Murphy shuts down the RBA’s reaction function near the effective lower bound, ignoring some important policy shifts, not least the introduction of QE in November 2020. This leads the model to both over-estimate and under-estimate the contribution of monetary policy at various points (as suggested by the Taylor rule residual from the baseline scenario). More seriously, it means that at least some of the contribution made by monetary policy is probably misattributed to fiscal policy. There is also very little discussion of the open economy aspects of the scenarios in the paper, although they are sitting in background. This goes to my critique of the RBA’s initial response to the pandemic, which is that it led to a macro policy mix that contributed to open economy crowding-out effects via the long-end of the yield curve (the short-end being held down by yield curve control). The AUD outperformed its G10 peers from March 2020, while Australian long bond yields were the highest in the AAA-rated sovereign credit space until QE came along in November 2020. They were even at a premium to the European periphery.
Chris’s paper is intended to simulate the effects of fiscal, not monetary policy, but the monetary policy reaction function should better reflect monetary policy developments over this period. The lessons Chris wants policymakers to conclude from his paper are: (1) don’t over-compensate; and (2) monetary policy should take more account of what is happening with fiscal policy. My concern is that policymakers will draw a different conclusion: that fiscal policy is more potent than it really is and should be used aggressively in response to shocks in preference to monetary policy. I have suggested to Chris that he also better simulate COVID-era monetary policy as another scenario as a way of better identifying the relative contributions of monetary and fiscal policy.
The NY Fed has released a study of Australia’s experience with yield curve control. It is somewhat telling that we need a foreign central bank to review our experience with this operating instrument. The finding that yield curve control had very narrow liquidity effects is consistent with my view that we would have been better served by a more broad-based QE program, in which a wider range of government and non-government securities were eligible to be purchased.
Last week, we noted that Australia underperformed both the OECD and G20 in attracting FDI inflows in 2021 and that it was yet another year of gross outflows for US FDI in Australia. Indeed, you might be surprised to learn there have been more Chinese than US FDI inflows every year from 2018-2021. For all the diplomatic frictions, the level of Chinese FDI has remained broadly steady with a modest 4.4% share, while the US share of inward FDI in Australia in level terms fell for sixth consecutive year to 17.4%.
More interesting than the gross inflows, however, are the gross outflows. Australia sent $11.4 billion in FDI to the US and a massive $89 billion in portfolio investment, including nearly $100 billion in portfolio equity securities. To put that in perspective, gross portfolio outflows for all countries were $162 billion. By contrast, we imported a mere $76.4 billion in portfolio equity capital from other countries, of which the US accounted for $69.3 billion. Australia is now a massive net exporter of portfolio equity capital.
The result is that, in level and absolute dollar terms, Australia is now a bigger investor in the US than the US is in Australia. The stock of Australian FDI in the US and US FDI in Australia is almost identical at $185 billion. Australian portfolio investors hold $78 billion more in US stocks and bonds than the US holds in Australia.
The drivers of these outcomes are discussed in my USSC report on the bilateral investment relationship from last year. But the straightforward implication of these data is that, in absolute dollar terms, Australia is now doing more to fund US investment than the US is doing to fund Australian investment. Australia’s status as a net capital exporter highlights its poor domestic investment performance.
ICYMI
Contrarian indicator of the week: Roubini working on an asset-backed digital currency to replace the US dollar, so of course the US dollar is making new two-year highs.
Your tweets: