Cees Bruggemans, Chief Economist of FNB, discusses the various factors that a central bank (such as the Reserve bank in South Africa) should consider when making interest rate decisions. He wrote the following article after the SA Reserve Bank to keep further interest rate cuts on hold earlier this month:
Interest rate decisions are not easy moments...
There are many aspects taken into account when deciding where to pitch interest rates. Given today’s fashions in the world at large the main considerations can be simply summarized and constitute perhaps a useful checklist for future occasions to signal rate changes.
The Taylor Rule, now about two decades old, tries to capture four thoughts. In addition, we also now have to take into account asset prices and risks to the forecast.
In terms of Taylor, firstly, any market-based capitalistic system with an active well-developed banking system at its core needs real interest rates to reflect the cost of money over time relative to generally changing price trends (inflation/deflation).
To discount the future to the present should cost something, if only to signal scarcity. This premium should be high enough to prevent excesses but low enough not to intimidate unduly. Similarly, saving resources in the present for the purpose of delayed consumption in the future should be worth a premium reward.
The Fed unofficially defines this real rate required as being 2% (for its intervention rate, our repo rate, over time). If banks maintain a 3.5% spread, a real prime rate of 5.5% comes into focus, something that we can observe over long periods of time in our own financial history.
Secondly, interest rates are one of the most important prices in the economy, with the exchange rate being the other major pivotal.
In an increasingly integrated world with free capital flows, a policymaker stands little change of controlling the exchange rate. Indeed, trying to do so (betting against market forces) can prove very costly.
For policymakers to influence the economy successfully, the preferred lever is short-term interest rates. Such policy interventions may at times be necessary as and when the economy starts to show signs of imbalances and disturbances, preventing optimal performance.
The central aim of policy in support of long-term growth is to maintain financial stability. To this end a stable real short-term interest rate is seen as anchoring the economy and expectations.
But things hardly ever remain stable. Shocks occur (from the outside) and human behaviour is temperamental (from the inside).
Both these forces have a way of deflecting economic growth from a stable potential trajectory (determined by long-run institutional features such as rate of fixed investment, supply of labour, rate of technological change, including qualitative changes in the labour force, regulatory limitations and policy interventions, for good or bad).
These same forces, and other influences, decide the extent to which prices are behaving, and any broad inflationary or deflationary tendencies.
Bearing all these aspects in mind, these many forces are on the loose daily, threatening the stability of real interest rates and thus the optimal growth trajectory, from time to time inviting policymakers to act and undertake course corrections.
Thus besides incorporating a real rate premium, interest rates should at least match (reflect/incorporate) the expected inflation rate to come.
In addition, there is embodied within Taylor the analysis and proposed treatment of two major possible deviations from stability inviting policymaker activism, namely deviating output tendencies (over- or undershooting potential growth) and changing inflationary conditions (over- or undershooting an optimal price trend).
Thus, thirdly, Taylor interrogates the inflationary condition. Is the inflation condition stable, near the ideal policy norm (2% in most industrial countries today, but in our instance still pitched at 3%-6%)?
Or is the inflation trend deviating, and changing future expectations, reinforcing this change and starting to influence the economic growth trajectory as well?
If deviating, and creating a so-called ‘inflation gap’, Taylor suggests that in order to dampen accelerating inflation behaviour beyond the policy norm, nominal interest rates should be raised sufficiently to undo such expected inflation acceleration (and its expectation).
This is done by ensuring that the real interest rate is being maintained. In other words, the expected rise in inflation over and above the policy norm is matched with a rise in nominal short-term interest rates, thereby exerting countervailing pressure to the disturbing influences emitted by the accelerating inflation signal, inviting it to subside back towards the policy norm.
Fourthly, the condition of the economy may also become disturbed. Growth may be under- or overshooting its potential, best described by an ‘output gap’ observed by comparing actual resource utilization with potential resource utilisation (a combination of labour force and physical production capacity utilisation).
To the extent that the economy is underperforming its potential, interest rates may be lowered temporarily. Similarly, if the economy is outperforming potential it could overheat, inviting temporary interest rate increases to temper such excess demand on resources.
So far, we have established an ideal policy setting, in which the economy is progressing along its potential growth trajectory accompanied by appropriately limited inflation reinforcing this ideal growth condition.
Guiding this condition is an appropriate real interest rate regime. If deviations occur within the inflation and/or the growth condition, countervailing action can be taken to raise, lower or keep nominal interest rates largely unchanged.
Over the years, however, refinements to this simple Taylor guide became necessary. One concerned asset prices. The other concerned risk (that things may not work out as thought).
Asset prices also can have a way of fundamentally influencing the condition shaping the economy’s growth trajectory. Exuberantly rising asset prices can invite excessive risk taking and can lead to overheated growth conditions because rising asset prices generate wealth effects that boost consumption expenditure. Similarly declining or excessively depressed and stagnant financial asset prices may inhibit risk-taking, thereby creating a drag effect on actual economic growth through negative wealth effects.
Whereas it is extremely difficult to judge market price signals as to whether asset prices are acting excessively exuberantly or depressed, it is nonetheless accepted today that there can be feedback loops to inflation and growth over time.
Lastly, every central bank is confronted with risk, that its readings of the future turn out to be wrong (wrong assumptions), or that new developments, unforeseen or otherwise, cause a different outcome from the one presumed in the policy stance at any moment in time.
As interest rate changes take up to two years to fully work in on the economy, central banks face potentially a two year window of risky uncertainty in which their original actions may or may not be derailed.
Therefore, after taking into account current views of the likely inflation and output gaps expected over the next two years, and the likely behaviour of asset prices (especially houses, equities and bonds), there comes a final moment where the central bank should second-guess its own main assumptions and bias.
If there is a clear risk formulation of potentially being wrong in a particular direction, this may encourage insurance being taken out in pitching the interest rate decision accordingly with some bias built in (out of safety, just in case).
The Americans call this over-engineering, but such safety precautions are taken for good reason. One rarely can foresee all future mishap and surprises, and to the extent that there is disagreement about the likely course of events, such sense of risk should be incorporated in any decision to a sensible degree.
One is now ready to take the plunge daily and set interest rates.
Because certain things change only slowly, and one does not want to confuse market participants unduly with frequent policy changes, a certain time lapse is advisable between policy deliberations about any course changes to be made.
Today’s decision probably had a few simple ingredients.
The CPI inflation rate is now 8% and is forecast to decline towards 5% later next year. The inflation gap is at least 3.5% (the difference between 8% and the midrange of the 3%-6% CPI target range).
Meanwhile the economy is in serious recession, this year probably registering 1.5% decline in GDP, even though the potential growth rate is closer to 3.5%.
Depending on how one wants to calculate potential relative to actual GDP, the output gap could already be in the 4% to 5% range across the broader economy.
Asset prices remain currently depressed or are still declining. Equity prices are at least 35% below their (overheated) peak of last year, though stabilizing with an upward bias, while price/earnings multiples today seem to be more realistically priced for the long term.
House prices will likely fall 10% in nominal terms peak-to-trough, but again from probably overheated levels, with at least another five years of eroding real house prices ahead of us.
Risk today mainly focuses on the depth of the recession and output gap (it could get somewhat bigger than expected), implying also downward pressure on inflation.
But core inflation also shows evidence of rigidity. Food price inflation is slow in coming down. Services inflation is high and sticky. Wage and salary trends are typically backward-looking, seeking compensation for recent inflation surges, besides demanding any premiums for reasons of skill scarcity, political influence or simply union size and strength.
In addition, there are future commodity price surges to worry about, oil having again risen by half this year and markets generally being worried about current central bank actions and future inflation playouts (such worries creating premiums, whether warranted or not).
Summarising all this requires some appeal to the Wisdom of Solomon. The main idea, though, is that the real interest rate premium of 5.5% is sacrosanct. We are expecting a 5% CPI inflation rate next year which needs to be matched by interest rates today.
That gives us a minimum prime of 10.5% before deviations and risks are to be addressed.
Here we have at least a 3.5% positive inflation gap and a 4%-5% negative output gap. Adding up and dividing by two (as is done in the real world) gives us a small negative 0.5% policy activism requirement, lowering the intended prime to 10%.
Asset prices look low and depressed and are possibly exposed to further decline, though there seems to be an upward bias globally. Yet when these current asset values are set off against recent overheated peaks, and taking into account cyclical behaviour and likely future income streams, these asset values don’t look overly out of place. Perhaps a slight accommodation bias may be warranted, but let us not overstate our case.
That leaves risk, all of it negative on the inflation side (as mentioned) but also on the output side, with the growth and employment loss possibly turning out bigger than expected, still undershooting over the coming year.
Is all that risk worth a 1% premium, the targeted prime rate rising to 11%, matching the ruling prime interest rate and warranting no change to today’s policy stance, with perhaps a rain cheque taken to look again at the situation at the next MPC meeting in August?
Or should we build in only a 0.5% risk premium, boosting the target prime rate back to only 10.5% and warranting a 0.5% rate cut from the ruling 11% prime to 10.5% today?
As to future movements in interest rates, keep track of the real rate at 5.5%, the CPI inflation 18 months out (how will it deviate from 5%?), the current inflation rate (it will fall from the current 8%, reducing the size of the inflation gap), the output gap (it will eventually stabilize and then narrow as recovery proceeds, but how fast will this in fact happen?), asset price behaviour (stabilization and gradual recovery, but nothing that needs more policy support?) and risks to the SARB policy forecast in every dimension.
No comments:
Post a Comment