In June 2006 most of us thought the ECB should not aim higher than 3.5 percent. In June 2007 almost all of us supported a move to four percent and many had a bias toward further hikes. What has changed?
I see essentially two reasons why our thinking has changed so much in the last 12 months.
First, global growth has surprised on the upside. This led monetary policy environment worldwide to be tighter than previously expected offering thus the ECB comfortable leeway to continue normalizing the level of its key rates without going against winds. With growth in the Euro area also supported by international conditions and remaining above its potential until now, the ability to face the euro appreciation proved to be higher than initially expected. All in all limits to ECB rate hike progressively disappeared.
Second, flattening or inverted yield curves in Western economies proved to be less negative on credit growth than initially thought. High competition between credit institutions and high liquidity from abroad might explain part of this inertia of credit growth to changing yield curve. One might also be tempted to see the absolute level of rates as having more impact in current circumstances on credit demand than the yield curve has on the supply side.
Those two points, i.e. international conditions and sensitivity of activity to the level of rates, are likely to remain key factors of the ECB policy background in the near future.
Last month, I defended the view that the ECB would have to tighten significantly more its monetary conditions in the future if current growth conditions were to remain intact. While sticking to this view, I do not see any reason to rush or accelerate the path of rate hikes. It is indeed important that rates move progressively if we want to preserve current economic conditions. Bond yields are also contributing the tighten credit conditions. This is a new trend that required to be closely watched to judge how much of the central bank job this trend is doing. The ECB raised its key rate last month to 4% and can now wait probably two or three months until next move without talking any risk on the inflation front. I then recommend no change in rate this month but keeping the tightening bias for the future.
There are signs that Euro area economic growth in the second quarter will slow somewhat. It is likely but by far not yet certain that economic growth will gain momentum again in the second half of the year. Moreover, inflation is not showing signs of picking up in coming months. For all these reasons I think a longer pause in rate hiking is appropriate. If economic growth should strengthen more and inflationary expectations increasing more in the course of the next six months than appears likely to me there is enough time to hike rates in October/November. According to my calculation the current policy rate is already at the upper margin of a neutral level (3.5%-4.0%) and for the time being there is no reason to move the policy rate into restrictive territory.
I see no hurry for a new hike. Inflation is well behaved and unit labour costs are flat. It is hard to be sure that potential growth has increased subatantially before we reach the peak. And the euro remains very high. Speeding up interest-rates hikes would deliver a strange message to G7 partners, especially Japan. Many economists would argue that the yen rather the euro is misaligned.
Neutrality is a strange concept. It's easy enough to define in theory but much more difficult to identify empirically. It's a little like the natural rate of unemployment. Although neutral rates may be close to the growth rate of nominal GDP (or nominal GDP per capita) over the very long run, I don't think this definition helps when it comes to the short-term determinants of good monetary policy. Neutral rates are often defined as those rates consistent with the lasting achievement of price stability but the actual level of rates that determines neutrality will tend to vary. For example, periods during which asset prices deliver inflated returns will require higher neutral rates than might otherwise be the case. Many people argued that Fed funds were suitably high in the late-1990s based on previous history but, relative to the extraordinary stock market returns at the time, they proved too low. For the ECB currently, it seems to me that the big concern must be a disconnect between monetary policy - as defined by the level of short-term interest rates - and monetary conditions - as defined by money supply and credit growth. Liquidity is increasingly being driven by events beyond the ECB's control - most obviously the impact of the carry trade linked to BoJ monetary policy and the build up of foreign exchange reserves within emerging markets. For this reason, I suspect standard measures of neutrality may under-estimate the required level of interest rates consistent with ongoing price stability and, therefore, there should be a bias towards higher rates.
We have to admit that we underestimated the momentum of the Euro-economy. The positive growth surprises in the course of last year and to date justify the upward adjustment concerning appropriate ECB key interest rate and the peak of the hiking cycle; all the more so as there were positive surprises on the labor market. Resource utilization in terms of both labor and capital has risen, indicating possible upward pressures on wages and prices.
Compared to one year ago the upward revisions in inflation forecasts are less pronounced than for economic growth. But nonetheless, the present forecasts indicate that inflation 2007 and 2008 could continue to slightly exceed the definition of price stability, as in the past seven years.
In this respect, it was warranted to raise rates higher than expected, at least to a neutral rate of around 4 %. Having reached this rate, it now seems to be justifiable to wait for the effects of this policy and to decide further steps with the incoming data. It is not yet clear, that restrictive territory needs to be entered.
Charles Wyplosz (II)
Two of my distinguish colleagues seem to establish a link between the potential GDP growth rate and the neutral interest rate. Would they care explain the underlying logic?
It is clear that the normalisation of long-term risk-free real interest rates we are witnessing has implications for the neutral rate of interest – the intercept term in the Taylor rule. The implications are not immediate, as the neutral real rate is the equilibrium real rate over the horizon of the inflation target implicit in the Taylor rule. Many economists appear to have a two-year to three-year horizon in mind. My own view is different: inflation is to be brought as close to target as possible from now until Kingdom Come. This would not in an of itself make the appropriate real rate the yield on an index-linked default-risk free consol, because the horizon of the objective function can be longer than the period over which today’s nominal interest rate decision exercises any influence on future inflation. That effectiveness horizon (weighted by the degree of effectiveness at each horizon) would be the appropriate maturity of the policy-relevant neutral real rate. Three to four years is my swift and imprecise guess. That neutral rate has gone up by at least 50 bps in recent months. I consider this increase in the neutral rate to be the result of an exogenous shock – the correction of an asset market anomaly – extraordinarily and excessively low risk-free real rates globally, and not the result, say, of a global reduction in the excess of ex-ante saving over ex-ante investment. That may will happen in the future, however, making for a further increase in risk-free real rates and in the neutral rate.
The current increase in the policy rate implied by this increase in the real neutral rate is likely to be less than the full 50 bps (say) increase in the level of the neutral rate, because the increase in the real neutral rate, and in real rates generally, has a depressing effect on aggregate demand and therefore on the output gap – another argument in the Taylor rule. In the long run, with a zero output gap and inflation on target, policy rates will of course have to rise by the full amount of the increase in the neutral rate. I therefore expect policy rates in Euroland to go on rising beyond the five percent level, to get into the restrictive zone where they still have not arrived.
Rates can no longer be described as very low and are moving into the range where they can be characterised as being neutral. Hence, we need to observe the incoming data more closely with regard to the effects of past interest rate increases and continuing economic strength and upside risks to price stability. Having just raised rates in June, not enough data have been published to come to a fresh assessment where the economy stands. My expectation would be that growth remains strong, probably requiring at least one more hike later this year. But this view needs to be checked against reality in coming months.
My main change in view with respect to 12 months ago is a perception of a higher rate of potential growth, which has lifted my perception of the neutral level of rates. Neutrality is a useful concept, for it anchors the dividing line between accommodation and restriction. Before I thought neutrality to be around 3.5-3.75, now I see it at around 4-4.25 as growth prospects have improved.
The European economy has reached a plateau at around potential growth rates and it is no longer accelerating. The EUROCOIN indicator is very clear in this respect and, together with a projection of growth at around potential, indicates that there is no reduction of slack in the forecast. The question therefore is about the amount of slack existing at the moment and, in my view, slack still exists, as the absence of inflationary pressures after a few quarters of strong growth reveals. The Phillips curve has flattened significantly and the Nairu has probably declined, in an environment of stronger prod performance and a bit higher potential growth, With unit labor costs flat, inflation well contained and slack still present, rates should be in the lower area of neutral, and that is where they are now. In my view, with potential growth rates at around 2-2.5, rates can stabilize at around 4-4.25 for a while, and therefore I do not foresee to vote for another rate increase for a few months. Ideally, after a series of 3-month rate hikes, I would now shift to 4 or 5 month intervals until reaching 4.25-4.5 percent, a point at which I would probably declare the hiking process over. I would also like to comment on the fact that the same way that a few years ago the perception of the euro area economy was too negative because of the weak situation in Germany, now it may be too positive because of the strong situation in Germany. We should not forget that the needs of Germany do not decide the monetary policy stance of the ECB. The Spanish housing is very weak and can weaken significantly from here, and other European housing markets are at risk too. At some point, if not already now, the ECB may have to follow what the Fed did last year and consider that there will be a time when rates will have to go on hold despite inflation risks being still to the upside, for the best policy will be to keep rates high for longer rather than rising them further. The focus on the monetary pillar and asset prices argues to be preemptive to then be able to stop sooner and prevent a clear bust. This hypothesis may be about to be tested.
In my view, the interest rate is now close to neutral, maybe still a bit on the expansionary side. Consequently, I anticipate the need for further increases. Last year, I was of the view that the peak would be at around 5%. This remains my central assumption today, with a possible 50bp margin error. On the down side, the euro appreciation is acting as a contractionary force and could lead us to a lower interest rate peak. I am also struck by wage moderation as exemplified by the recent Deutsche Telekom agreement. On the up side, along with sustained growth in Asia, a rapid return to growth in the US could further invigorate European exports and bring wage moderation to an end, possibly quite swiftly.
At just over 4%, short-term interest rates appear well in the range of estimates of neutral. Past rate increases have moderated narrow money aggregates and credit demand by households substantially, as well as house prices. Credit demand by firms remains very dynamic, although roughly half of it reflects financing of share buy-backs and M&A rather than real spending. The recent back-up in medium- and long-term yields implies a meaningful tightening of monetary conditions, similar to an equivalent increase in short-term rates. As a result, the need for higher short-term rates is further diminished
GDP growth appears to have stabilized at an annualized pace of between 2% and 2.5%, that is, very close to current estimates of trend, and the outlook for the coming quarters points to a similar performance. Inflation remains anchored just below 2%, that is, in line with the target. That result implies, in my view, that underlying price pressures are contained, as suggested also by the fact that core inflation (excluding the impact of the extraordinary indirect tax increases this year) remains stuck around 1½% so far this year.
My own estimate of the euro area's potential growth rate is creeping up. So my own estimate of the neutral rate is also creeping up. Hence I am more comfortable recommending rate hikes for the ECB because at this stage of the business cycle I would rather err on the high side than vice versa. High and increasing asset valuations reinforce the idea that one can minimize the risk of asset prices overshooting their equilibrium by increasing rates.
On inflation: we see for the time being the usual pattern continuing: inflation is about 0.5 % above the level forecast a year earlier. Thus, I would expect inflation next year to be slightly above 2 %. No reason to be alarmed, but to be cautious.So, in my view, at this point in time all argument point to higher rates.
This is what we thought in June 2006
(from the minutes):
June 1, 2006: There was a discussion on the committee about the peak of the interest rate at which the ECB should aim, based on current information and expectations. Charles Wyplosz argued that getting to five percent within six quarters would seem appropriate. Most members thought this target was much too high. The majority thought that 3.25 to 3.5 percent was an appropriate peak to aim at for the current cycle. Charles Wyplosz argued that it was normal to expect that the policy stance would have to move from expansionary to restrictive at some point rather than just going to neutral and staying there. Philip Lane provided some support by arguing that economic developments in Asia made it likely that this region would not keep global price pressures down as much in the future as in the past. Most members however felt that going to neutral would be enough, given that wage pressures were regarded as very muted and only a moderate economic upswing was expected.
June 29, 2006: Willem Buiter weighed in to side with Charles Wyplosz and Philip Lane who had argued before that the majority estimate of about 3.25 to 3.5 percent was too low. Buiter put his own estimate of the neutral rate at above 4 percent.
Note: The consensus growth forecast for 2007 at the time was 2.1 % (currently 2.6 %,) inflation forecast 1.7 % (currently 2.0 %).