Waiting for ECB to detail its exit strategy

• ECB. When the Governing Council convenes next Thursday, interest will be focused not on the quarterly update of the ECB macro forecasts but rather on the details of the exit strategy. The central bank already outlined this at the end of last year.

• Refinancing. We expect the ECB to announce the return to the pre-crisis framework: Main refinancing operations will be carried out via (full allocation of) the 1W tenders, fine tuning via the 3M operations, which will be "resubjected" to a competitive bidding process. The special tenders will expire (pages 5-6).

• Interest rates. We still do not expect a refi rate hike until the beginning of 2011. But as excess liquidity will successively be removed, money market rates, which are currently well below 1%, should gradually converge towards this key rate level in the course of the second half of this year.

• Outlook. This is based on the assumption that the ECB has no need to make any major adjustments to its macro projections. Despite brisk global trade and the depreciation of the euro, the EMU-wide economic recovery will be restrained in 2010/11 – not least because of the need for further consolidation in periphery.

• Further topics:

  – Weekly Comment: Avoiding exit-phobia (page 2).

  – Germany: A slightly different labor market cycle this time (page 7).

  – Italy: Pension system in relatively good shape, but ... (page 10).

  – US: Pain in commercial construction to remain (page 12).

  – Oil price: Short-term forecast models & the financial crisis (page 15)

  – Data outlook: Rise in inflation in the eurozone takes a breather; higher US job losses despite census hiring (page 19).

  – Market outlook: Euro to remain vulnerable (page 28).

Exit-phobia

We all agree that the extraordinary and unprecedented policy stimulus rolled out in response to the crisis will eventually have to be unwound. But there is much less agreement on when the unwinding should start, and the temptation is to say, not yet. The IMF has published a comprehensive examination of the exit problem Exiting from crisis intervention policies where it also argues that on balance the risks of an early withdrawal of stimulus dominate those of a late withdrawal. I have been a vocal advocate of early and decisive intervention by monetary and fiscal authorities during the crisis, but I am now feeling increasingly uncomfortable with some of the arguments for a late exit. In particular, I am concerned that the exit strategy might become open-ended, and that prolonged policy stimulus might divert attention and pressure from the need to put in place the conditions for sustained private sector growth. This is dramatically apparent for fiscal policy, which needs to exit not just the anti-crisis stimulus, but a longer-term expansion path that is clearly unsustainable, even without factoring in the cost of population aging. The fiscal adjustment needed to bring the advanced countries’ debt to GDP ratio back to 60% is daunting – but a smaller or slower adjustment would come at the cost of lower economic growth over the long term. It might seem ungrateful to criticize loose fiscal policy after begging governments to save us from depression – but the harsh truth is that high government debt absorbs precious private savings and would cause higher long-term interest rates, making financing to the private sector more expensive and scarce. On the monetary policy front, the exit strategy is a three-headed monster, requiring decisions on liquidity measures, policy rates, and asset purchases. The Fed is gradually unwinding some of the exceptional liquidity measures, even as it reiterates its pledge to keep the Fed funds rate exceptionally low for an extended period – and the disposal of purchased assets is not even on the table for now. The ECB next week should also announce its next steps towards unwinding exceptional measures and shortening the duration of outstanding liquidity. In particular, it should switch the 3-month refinancing operations back to a competitive auction, and charge a variable rate or a spread on the last 6-month operation. The sluggishness of the recovery and the fragility of parts of the financial sector are valid concerns, but they are largely structural. The ECB should not (and I believe will not) rush to tighten policy faster than the economy can withstand – but neither should it be tasked with keeping parts of the financial system on permanent life support. History warns us against the risks of premature policy tightening – but it also warns us against the risk of keeping zombie institutions around. Bottom line: policy stimulus should be withdrawn gradually and carefully, but we should not fall prey to exit-phobia.

The main argument for delaying policy tightening is that the economic recovery remains sluggish and uncertain, and the financial sector is still fragile. Policy stimulus therefore should be withdrawn only once there is sufficient evidence that private demand and investment have recovered momentum. I have no illusions on the strength of the current economic recovery, or on the resilience of the financial sector. I do think, however, that some of the weaknesses we are facing are structural, and it is therefore not clear at all that they would be best addressed with a prolonged cyclical stimulus from monetary and fiscal policy. Moreover, I think we should also carefully consider the risks that prolonged policy stimulus may adversely affect the longer-term growth potential of the private sector.

This is perhaps especially obvious on the fiscal policy front. Most governments, especially in advanced economies, are planning to keep stimulus measures in place throughout this year – although in some cases, market pressure for an earlier consolidation is increasing. Yet the fiscal consolidation challenge is daunting, and the IMF’s paper makes it painfully clear. According to the IMF, in order to reduce the debt to GDP ratio back below 60% by 2030, advanced economies would need to improve their structural primary balance by 8% of GDP, raising it from a deficit of 4 1/3 % to a surplus of 3 2/3 % by 2020, and then keeping it at this higher level for ten years. That means 8% of GDP of cuts in non-interest expenditures and/or tax increases; and that would not address the impact of population aging, which would require an additional 4-5% of GDP in corrective measures, ideally targeted to prevent or at least limit the rise in pension and health spending.

But what is perhaps most striking is that fiscal policy does not really need to exit from the anti-crisis stimulus, it needs to exit from a longer-term expansionary path that is clearly unsustainable, and that has been exacerbated by the discrete jump in debt ratios triggered by the crisis. The IMF notes that unwinding the fiscal stimulus would contribute 1½% of GDP of the 8% adjustment required. Fiscal balances have of course deteriorated by much more than 1½% due to the recession, and they will improve significantly more as growth recovers, but that reflects the impact of automatic stabilizers and fluctuations in growth, which are netted out of the structural balance. The bottom line is that governments in advanced economies would need to strengthen their fiscal balances by 6½ % of GDP over and above the unwinding of the discretionary fiscal stimulus of the last two years.

The required fiscal adjustment would of course be smaller if one adopted a less ambitious debt/GDP target, but the costs in terms of lower economic growth would be significant. Rather than aiming to return to 60%, the median 2007 debt ratio for advanced economies, we could settle for stabilizing the ratio close to 100%, which we are projected to approach over the next few years. A quick look at Italy and Japan, however, provides a sobering reminder that stabilizing the debt/GDP ratio at high levels does not seem conducive to a robust growth performance. It might seem ungrateful to criticize loose fiscal policy after begging governments to save us from depression – but the harsh truth is that in the long run high levels of government debt absorb precious private savings and would likely cause higher long-term interest rates, making financing to the private sector more expensive and scarce. The IMF estimates that long-term interest rates rise by 5 basis points for every additional 1 percentage point in the debt to GDP ratio; that implies that the difference between 60% and 100% debt/GDP comes to a permanent increase of 2 percentage points on interest rates. ((For an even harsher view
and some truly unpleasant fiscal arithmetic, Robert Barro’s latest WSJ piece is definitely worth reading). The IMF’s “exit paper” also warns that “…a moderate increase in inflation to erode the real value of debt would not make much difference to debt trends but would de-anchor inflation expectations and eventually entail large output costs to restore price stability.” This is notable as the IMF had attracted some criticism for suggesting that central banks could perhaps target a slightly higher inflation rate. While the IMF’s argument was that this would simply leave more room to cut rates in an emergency, markets immediately suspected a not-so-hidden agenda to engineer a stealth reduction in public debts via higher inflation. The statement quoted above suggests such suspicions are probably mis-directed.

Interestingly, it is monetary policy and not fiscal policy that has been accused of having been too loose for too long, thereby helping to lay the basis for the crisis. So where do central banks stand? The exit strategy on the monetary policy front is a three-headed monster: central banks need to consider (1) how to unwind liquidity measures; (2) how to unwind purchases of public and private sector securities; and (3) how to normalize policy rates; the three heads can move at different speeds and conceivably even turn in different directions.

The Fed generated some excitement last week when it raised the discount rate by 25bp, triggering speculation that this might be the prelude to an earlier rise in the Fed funds rate. As my colleague Harm Bandholz pointed out (“How to interpret the increase in the discount rate?”, 19 February), however, there is no automatic connection between the two, and the increase in the discount rate represents a reaction to the ongoing normalization in financial conditions rather than a discretionary step to tighten monetary policy. Lending at the discount window has diminished substantially, to about USD14bn in mid-February from about USD 110bn in the aftermath of Lehman’s collapse. This suggests that the Fed can safely move to normalize the spread between the discount rate and the Fed funds rate, bringing it gradually back to 100bp, without an adverse impact on market conditions. Of course, the fact that financial market conditions are improving means that we are getting closer to the point where the Fed will feel sufficiently confident to increase the Fed funds rate, but before we actually get there the Fed will need further evidence that the recovery is sustainable. Fed Chairman Bernanke on Wednesday reiterated the pledge to keep the Fed funds rate at an exceptionally low level for an extended period, which decoded means we are at least another six months away from the first rate hike . We still expect that the Fed will start hiking rates before the end of the year, although we see some risk to our September call for the first move. Meanwhile, the Fed will continue with the gradual normalization of liquidity conditions. Unwinding the purchases of assets is much more difficult, given the risk of adverse repercussions on the yield curve and in particular on mortgage financing costs, and will therefore be left for the very end of the adjustment process.

The ECB faces an important test next week, when it should announce the next steps in its exit strategy. The action is all on the liquidity measures: the ECB’s direct purchases of assets have been negligible, and I expect the refi rate to remain on hold throughout this year. The ECB has already discontinued the 12-month Long Term Refinancing Operations (LTRO) and announced that the next 6-month LTRO will be the last one. The key question for next week is whether the 3-month LTROs and the last 6-month LTRO (to be held on 31 March) will still be conducted with full allotment at a fixed rate (1.0%).

I think the ECB should and will switch to conducting 3-month LTROs as competitive auctions, and apply a variable rate or a spread to the last 6-month LTRO, while keeping the regular 1-week Main Refinancing Operations (MRO) at full allotment, fixed rate for the time being. This would allow the ECB to shorten the maturity of the outstanding liquidity, which could then be withdrawn more quickly if needed. There is concern that these steps might be premature given the sluggishness of the recovery and the fragility of the financial sector. But as I argued at the outset, these problems are largely structural. We have been highlighting for some time the dichotomy in the eurozone’s financial system, where some institutions have regained near-normal access to market funding, whereas others remain dependent on the ECB. But if some financial institutions face a fundamental problem because their business model is no longer viable, the ECB should not be charged with keeping them afloat –that should be a problem for regulators and governments. Indeed, past experience should warn us as much against keeping zombie institutions around as against risking a premature tightening of policies.

We no longer face the systemic dislocation and instability which characterized the worst period of the crisis, and the ECB should therefore gradually phase out the exceptional support measures deployed, nudging short-term market rates back in line with the 1.0% refi by year-end. If the ECB maintains the full allotment fixed rate procedure, we will probably see most of the liquidity that expires on 1 July with the first 12-month LTRO being rolled over at 3 months. But then at the end of September, we will have the simultaneous maturity of a large amount of liquidity from the second 12-month LTRO, the last 6-month LTRO (to be held on 31 March) and the 3-month LTRO to be settled on 1 July. The ECB would de facto be postponing the start of the exit strategy to September, and postponing to that date the problem of smoothing out the liquidity drain. Moreover, at that point the ECB would face another uncomfortable choice: whether to signal that the refi will remain on hold well into 2011, or to risk an abrupt upward adjustment in short-term rates. I think that while the withdrawal of liquidity will need to be gradual and careful, it should not be postponed all the way to September.

•  Download Full Friday Notes