Via Damien Boey at Credit Suisse:
We model the slope of the real yield curve – the spread between 10-year inflation-indexed bond yields, and real 3-month interbank rates. We have demonstrated previously that the slope of the real yield curve can be reliably modelled as a function of four factors:
- Home-buying sentiment.
- The forward-looking components of business confidence.
- The inverse of the output gap.
- The expected change in the US trade deficit, a proxy for the availability of USDs in offshore markets.
This model was pointing to extreme flattening, and even inversion in 2018, through to mid-2019. This flattening and inversion materialized. Now, the model is pointing to steepening, to the tune of 35bps. We arrive at a steepening signal, despite evidence that spare capacity in the economy has tightened, consistent with higher (rather than lower) rates, and the re-surfacing of a global USD shortage. This is because housing sentiment has dramatically improved in the wake of the LNP’s Federal election victory, credit easing, and several rate cuts. Also, it is because business confidence may be on the verge of bottoming out, because what firms feel about the macro environment has moved ahead of what they are currently experiencing across new orders, profitability, employment, inventories and the like. In other words, the market should be pricing in higher, rather than lower rates from here, because policy easing to date has been surprisingly effective.
It is also worthwhile mentioning that interbank credit spreads have plummeted this year, reducing the risk of out-of-cycle rate hikes from the banks, and impaired transmission. Therefore, the RBA has to do less work to achieve system stability, and that the neutral rate of interest is higher, rather than lower. We estimate that the long-term neutral rate currently sits somewhere in between 1.6-1.9%, compared with current 10-year bond yields of only 1%. There is significant room for a long-end sell off, and curve steepening on this basis alone. Of course, this does not feature explicitly in our modelling, because we have taken into account potential impairment to the monetary transmission mechanism through our measure of the short rate in the yield curve (3-month interbank rate), rather than through our estimate of the long-term neutral rate (where we assume for simplicity, full pass through).
Overall, we see no compelling case for more RBA rate cuts, based on the evidence that curve steepening is likely. But the RBA and money market investors have convinced themselves that more cuts are required, and imminently. Should the RBA cut rates today, historians will be able to pinpoint today as the exact moment that officials abandoned its macro-prudential mandate, preferring instead, growth at any cost. The RBA will officially have taken political ownership of whatever happens next in the housing market. There will be no one left to outsource blame to.
Does the Bank care about any of this? Even if the Bank should, perhaps it no longer does. Recent citation of the “global saving glut” as a reason for low interest rates everywhere (even though in theory there can be no such thing) and the lowering of the natural rate of unemployment down to 4.5% (despite evidence of accelerating wage inflation at an unemployment rate well above this level) suggest to us that Bank officials are looking for narratives to support cutting, having spent several years defending macro-prudential policy, and keeping its powder dry. They seem very concerned about undershooting risk in the economy, even though the undershooting that is occurring today is merely technical payback for overshooting in 2016-17, because in 2016-17, we saw the confluence of a foreign demand-driven residential construction boom, and an extreme pull forward of infrastructure spending. Indeed, there is very little that the RBA can do today to immediately reverse the undershooting that is coming. But they can limit and have already limited the fallout from growth undershooting on household balance sheets, by easing financial conditions.
As likely as capitulation seems today, we doubt that Bank officials really want to exhaust their ammunition, despite apparently advanced thinking about unconventional policy. Deep down, they know that unconventional policy does not really work, but for exchange rate devaluation. Targeted long-term repo operations merely keep troubled banks alive to be able to keep lending. Quantitative easing floods the system with reserves, and drops bond yields – but more reserves do not cause banks to increase lending, and no-one borrows at long-term interest rates. So asset purchases do not really relax any binding constraint on the economy. Exchange rate devaluation is more promising – but it requires the Bank to walk back in time to the pre-float era, having already walked back in time to the pre-deregulation era through macro-prudential policy. It requires the RBA to expand its balance sheet and effectively fight the Fed – a short-term strategy at best.
From a bond pricing perspective, another RBA rate cut today will only further the case for steepening, because the more (debatable) cuts that occur now, the less the need for future cuts. On the flipside, a pause today will also further the case for steepening, in the sense that the Bank can signal that there is no urgent need for extreme easing, because it is willing to back the “gentle” inflection point upwards in the economy. We think that money market futures are overpriced for rate cuts this year.
A few points on why I think Mr Boey’s model is breaking down:
- the RBA can (and has) passed the housing buck to APRA so it has political coverage;
- the RBA has practically hit the lower bond and is mulling QE, including sitting in on long end bonds;
- better credit conditions are not going to translate quickly into higher growth or inflation owing to over-leverage, the apartment defect crisis and mass immigration sitting on wages;
- the US trade deficit is not lifting global activity owing to the trade war, and
- a big fall in Australian income is underway and worsening as a commodity supply side shock unwinds.
All of these are good qualitative reasons for why the RBA is still cutting and ducking the message sent by a quantitative model. That is not to say that the leading credit index is no longer useful and we have seen some curve steepening. But these are good reasons to expect the FCI to at least partially dislocate from activity this time around.