Thursday 22 January 2015

ECB's QE: More Magic Than Mechanics

So, will it work? 

It would be easier to answer that question if we could form a clear idea of what Mr Draghi thinks this quantitative easing will achieve and how it will achieve it. That is not easy.

It seems to me that throughout this financial crisis there have been four models about what QE might achieve:

  • The first is relatively straightforward: quantitative easing has been (in the US and UK particularly) a way of publicly guaranteeing the solvency of potentially distressed financial systems. 
  • In the second model the central bank hopes to master the longer-end of the bond market, driving investors are driven into riskier instruments, and thus driving down risk premia. In terms of the economy, by depressing bond yields below 'fair value' rates, central banks and economists entertain the hope that savings rates would be cut, and investment spending encouraged. In both the US and UK this has happened very slowly, very late in a business cycle, and to the extent that it has happened at all there is no certainty (and limited probability) that QE played a key role in changing savings/investment choices. 
  • The third model involves using QE to announce a public 'regime change' of monetary policy which, by itself, manages to raise inflationary expectations. 
  • The fourth model is quite different and the polite financial community pretends it hasn't noticed it: in Japan, QE is being used as a way in which the central bank can achieve hegemony over/functionally replace a banking system which seemingly cannot be revived from its decades-long coma. 

Which of these engines does the ECB think it has set in motion?

The first move is to listen to what Mr Draghi had to say. The key passage, it seems to me was this: “while the monetary policy measures adopted between June and September last year resulted in a material improvement in terms of financial market prices, this was not the case for the quantitative results.” What does he mean by quantitative results? He could mean either there was insufficient positive results in terms of credit (and he's right, bank lending to the private sector fell 1.4%, or by Eu151 bn over the 12m to Nov), or alternatively, the 'quantitative result' he may be referring to could be economic output and economic growth.

Nor did Mr Draghi get significantly more coherent as he outlined what he thought might be achieved: QE would

  1. decisively underpin inflationary expectations 
  2. ease financing conditions for firms and households 
  3. 'reinforce the fact that there are significant and increasing differences in the monetary policy cycle between major advanced economies.' 

That last is simply obscure: the obvious explanation is that he is simply talking down the Euro – is this really what he intended?

If from all this you can construct a clear set of aims, and a picture of the mechanisms by which the Eu60bn per month buying of assets will achieve those aims, you are one step ahead of me. But if pushed, I would say he is relying on a 'regime change' to push up inflationary expectations, whilst hoping that ECB's bond-buying will somehow be passed on to firms and households.

There are two problems getting in the way of that second hope. The first, of course, is that the longstanding expectation that ECB would eventually be driven to something like QE has already depressed both sovereign Eurozone bond yields, and risk premia. Ten-year Eurozone sovereign bond yields are only around 50bps, with the risk premium of 10yr BBB bonds approximately 100bps, and, for troubled sovereigns such as Spain, around 140bps. The marginal impact of squeezing these premia down further can surely be only slight.
The second problem is that even as national central banks buy bonds from their own financial system, the receipts are likely to pool in the most credit-worthy systems. Within the Eurozone, that means the German banking system, where we will be able to track the process by the Bundesbank's Target 2 balance with the ECB. Outside the system, the Swiss National Bank is making a radical assumption that plenty of the ECB's QE is coming its way. For evidence, consider changes in the ECB's balance sheet:  


and the way the fluctuations of 2011 to 2014 have been mirrored in the Bundesbank's Target 2 balances with ECB: 


and the short term liabilities and foreign investments build-up in the Swiss National Bank:
Why should we expect it to be different this time?

Tuesday 20 January 2015

China in 4Q: Tactical Reverse Delivers Modest Victory

China's release today of December's monthly data and 4Q GDP results contain enough information to give us answers to two distinct questions:

  • To what extent have China's authorities succeeded in halting the slide of the first three quarters?
  • How much progress has been made in 2014 towards steering China towards a less resource-inefficient model of growth?

The answer is that it's reasonable to believe the slowdown in China's economy was indeed brought under control and in some respects reversed. But there is a price: virtually all measurements suggest China made no progress at all in 2014 in discovering a more efficient growth-model.

December Data and 4Q Growth

There was just enough in December's industrial and domestic demand data to suggest the underlying loss of momentum continued to moderate. For the industrial sector, the 7.9% yoy rise in output was a surprise exactly big enough to offset the fall to 7.2% recorded in November.  It included a 2.6% yoy rise in electricity generation which also just about kept that indicator conforming to trend.  Similarly, the 9.5% yoy growth in US dollar exports (9.8% yoy in Rmb terms, and 9.4% in volume terms) was very modestly greater than historic seasonal trends would expect. So the industrial sector ended 2014 in much the same state as it has been since 2012 - oscillating in a narrow range around, and usually just under, trend momentum.

Domestic demand has been the greater challenge as, broadly speaking, it tracked fluctuations in monetary conditions.  And December's data was collectively strong enough to show positive momentum for the first time in a year, which pulled up the 6m trendline slightly, although it is still solidly negative. The strongest signal was from car sales, which rose 16% yoy and were 1.5SDs above trend. In addition, retail sales growth of 11.9% yoy was 0.4SDs above trend for a second successive month. But these gains were offset by still-slowing urban investment (15.7% yoy ytd), and the continuing deterioration in employment conditions as tracked by the official manufacturing PMI.


My momentum indicators suggest that the deterioration of 1H has been mildly but successfully reversed in 2H, and particularly in the last quarter. And,  perhaps surprisingly, the quarterly nominal GDP results suggest the same thing.  This is not immediately obvious: nominal GDP growth slowed to 7.8% yoy in 4Q from 8.5% yoy in 3Q.  When one strips out the impact of the trade surplus (Rmb 917.3bn in 4Q14 vs Rmb554.3bn) in 4Q13 in order to get an idea of domestic demand, nominal GDP growth actually accelerated very mildly, to 6.1% yoy in 4Q from 5.7% in 3Q14.  Going further, one can also strip out the fiscal position, to get closer to movements in private domestic demand: we have the fiscal data only for October and November, but  judging from those two months, it seems clear that, despite the public commitment to supporting economic growth, the fiscal position actually tightened slightly during the quarter. (In the 3m to Nov, revenues rose 8.4% yoy and spending rose only 1.9%, and the Rmb 350.8bn deficit compared to a deficit of Rmb 540bn in the same period 2013).  As a result, when you exclude the impact both of the net trade position and the fiscal position,  I estimate the remaining private sector domestic demand grew 8.2% yoy in 4Q, up from 5.2% in 3Q and 5.8% in 2Q. In short, the deterioration was checked in 4Q.

That conclusion is also supported by my proxy for  the private sector savings surplus, comprising the trade surplus minus the fiscal position.  As the chart shows, the huge build-up of private savings surpluses which accompanied the slowdown throughout most of 2014, stabilized during the last few months of 2014, as confidence stabilized enough to cap the rise in precautionary saving. We do not yet have current account data for 4Q, but during the 12m to September, the PSSS rose to 4.6% of GDP from 3.5% in 3Q13.  For the time being, it seems likely that the ratio did not rise more in 4Q14.


Structural Issues - The Challenge Ducked

The evidence suggests that the government's attempts to avert a spiralling slowdown met with modest success during the latter part of 2014. But there has been a price: there has been no obvious sign that China is edging towards a more resource-efficient growth model. Rather, the longer-term deterioration has continued,  with the marginal improvements since the middle of 2013 scuppered in 4Q14.

My return on capital directional indicator expresses nominal GDP as a flow of income from a nominal stock of fixed capital, and I calculate movements in that capital stock by depreciating nominal gross fixed capital formation over a 10yr period. We do not yet have the formal by-expenditure breakdown of GDP for 2014, so 2014's 7.7% yoy investment spending is modelled from the 15.7% yoy rise in urban fixed asset investment.  This may prove a conservative estimate of investment spending in the national accounts, but even so, it implies China's capital stock is growing around 13.3% yoy - far faster than the c8.2% growth in nominal GDP. As a result, there is absolutely no sign that the fall in the directional indicator is easing up.
Perhaps it might be argued that after the huge investment frenzy of the last 20 years, it is quite unreasonable to expect a rapid turnaround in this indicator. However, it is difficult to see any improvement in other indicators, such as monetary velocity (GDP/M2): although the pace of deterioration has clearly moderated, it has probably not yet improved. (Monetary velocity may be interpreted as indicating changes in marginal output/capital ratios once the effect of changes in the credit cycle are accounted for.)

More directly, one can look at the economic efficiency of finance, tracking how nominal GDP has reacted to the addition of 1 yuan of bank lending, or more broadly 1 yuan of neg aggregate financing. At the beginning of 2014 there were signs that this was finally beginning to recover from the falls of 2008-2009 and 2012. However, developments in 4Q appear to have snuffed out that recovery: in 2014 one yuan of bank lending was associated with just 0.53 yuan of GDP growth, down from 0.77 yuan in 4Q13, and dipping back to the lows of early 2013.  Calculating the similar ratio for aggregate financing, one yuan of aggregate financing was associated with marginal GDP growth of just 0.30 yuan in 2014, down from 0.40 yuan in 2013.


None of this is to write the obituary on China's efforts to re-cast its growth model. But such a traverse is tremendously difficult at the best of times, and in 2014 China's authorities evidently discovered this was not the best of times. The economic strategy no doubt remains, but 2014 was a year in which economic tactics took precedence.

Sunday 4 January 2015

If ECB Goes QE, Remember Bundesbank's T2

Even if Mario Draghi can retain unanimous monetary policy board consent to a really sizeable programme of quantitative easing, the underlying untreated fractures in the Eurozone’s financial system make it difficult to believe it could significantly deflect the Eurozone’s economy much from its current trajectory. Put baldly, the mechanism by which the central bank can hope to transmit monetary policy initiatives throughout the economy are broken. And they are most broken where they are most needed.

Regardless of the ECB’s public policy pronouncements, movements in its balance sheet reveals what policy has actually been. And that policy has been to claw back the support to the Eurozone financial system it provided during the first phases of the Eurozone crisis in 2012. Between 3Q12 and the end of 2014, the ECB’s total balance sheet contracted by just over Eu1tr.  In calendar 2014, the ECB’s balance sheet shrank by Eu251bn, a contraction equivalent to approximately 3% of Eurozone GDP.

In terms of net lending to the Eurozone’s financial institutions, the total has fallen from roughly Eu650bn in 2H2012 to around Eu480bn in 2H14.  Evidently, the desire to shrink the ECB's balance sheet was a higher priority than steering the Eurozone away from deflation, fostering growth or eroding the unemployment totals of Southern Europe.

This will at least please the Bundesbank, since the fractures in the Eurozone banking system have forced it to become a massive lender to the ECB.   The problem is that Germany’s role as the Eurozone’s principal banking system safe haven results in large chunks of Eurozone liquidity pooling into Germany's banking system, which in turn results in the Bundesbank being the chief re-cycler of those funds back to the ECB. The chief tracker of Germany’s save haven/capital recycler role within the Eurozone is the fluctuations of the Bundesbank’s Target 2 net position with the ECB.  When Euro liquidity was fleeing Southern Europe financial systems and washing up in Germany’s commercial banks, the Target 2 net position of the Bundesbank with the ECB rose to a peak of Eu751bn in August 2012.  This was an amount equivalent to just under 25% of the ECB’s total balance sheet. Subsequently, this flow modestly reversed, with the Target 2 total falling to Eu470bn by March 2014, but since ECB also shrank its total balance sheet during the same period, the Bundesbank’s net position is still equivalent to 22% of the ECB’s total assets.  


Since March, however, the position has been largely unchanged, though over the past few months it has expanded very slightly. But this stability is not a return to 'normality': between 2000 and 2007, and prior to the Eurozone debt crisis, the Bundesbank's Target 2 balance averaged under Eu10bn. 

Now consider the implications of the relationship between movements in the size of the ECB’s overall balance sheet and the Bundesbank’s Target 2 balances with the ECB: they rise together than more recently have fallen together, but whilst the ECB’s balance sheet has returned to 2010 levels, the Bundesbank’s Target 2 balances are approximately two and half times what they were in 2010.  

What this tells us is that, despite what the fall in sovereign risk premia may assert, the perceived imbalance of risk in banking systems between Germany and the rest of the Eurozone has not been eradicated.  ECB’s guarantees of liquidity have suppressed risk premia, so that at present, Spanish 10yr sovereigns carry a bare 88bp risk premium, but if that premia has been ‘artificially’ suppressed by central bank actions and/or promises of action, it merely means that investors are no longer paid enough to offset the residual financial system risk. Hence liquidity continues to flow out of the Eurozone’s riskier banking systems and back into Germany’s banking system. 

The underlying fracture in the Eurozone between Germany and the rest of the Eurozone has not mended. The analogy of the ECB using a sticking plaster to treat a fracture is compelling: the smooth surface masks terrible and possibly irreparable damage beneath the skin. 

In particular, it illustrates just how limited any ECB ‘quantitative easing’ must be in effect, even if Germany’s representatives should allowed a concerted effort in that direction. For the evidence suggests that if ECB poured liquidity en masse into the Eurozone’s banking system, the economic and financial fractures in the Eurozone would result in liquidity quickly circling back once again,  quite uselessly, into Germany’s banking system.  Whilst this might – only might – help inflate German asset prices, it can hardly be expected to do the same for, say, Spain CPI, or Italian unemployment, or ex-German Eurozone growth. The underlying divergence directly sabotages the mechanism by which any conceivable (ie, nationally non-specific) monetary policy can take effect. 

It has been claimed that central bank quantitative easing can or has achieved different things at different times, using different mechanisms. The two most common beliefs are that sufficiently aggressive central bank intervention can effect ‘regime change’ which effectively encourages nervous financial systems with depleted risk capital, to reassess likely future returns and expand balance sheets which would otherwise be frozen.   Secondly, it has been asserted that if central banks can crush the risk premium across the range of financial assets, it can drive investors back into those ‘riskier’ assets from which they had recently fled.  

Both make the assumption that although balance sheets may be compromised and risk capital in short supply, the fundamental banking mechanism through which a central bank can act remains sufficiently intact to be rescued.  But in the Eurozone’s case, the enduring size of the Bundesbank’s Target 2 balance tell us that is not the case: Europe has many distinct national banking systems  with different risk characteristics masquerading as a single system.  But it's the still-giant Bundesbank Target 2 balances which reveal the truth.