Tuesday, 24 September 2013

Banking Union: a powerful reform

A single policy or reform can rarely achieve multiple objectives simultaneously. The Banking Union is potentially such a policy. First, the Banking Union can solve the credit supply malfunctioning within the euro area, namely it can permit the monetary policy impulse to be transmitted more uniformly across the Union. Second, the Banking Union can solve the credit demand malfunctioning within the euro due to the uncertainty on the future existence of the eurozone.

A short narrative

Consider Figure 1 that shows on the left panel the 1-month Libor on the euro (and the dollar). The crises in the euro periphery that resulted in the sequential bailouts program of Greece, Ireland and Portugal are closely related to the increasing tension in the european interbank market. The right panel of Figure 1 plots the spread between the Italian and the German bonds and shows that shortly after Portugal's bailout the tension precipitated in a fully fledged financial crisis. Figure 1 also shows the approximate dates of the ECB main interventions. These interventions have been highly successful in lowering tensions in the interbank market, first by intervening directly in the banking sector and second by announcing the OMT.  Figure 2 shows fixed investment spending for Germany, France and Italy together with the MFI interest rates on loans on new businesses. Fixed investment is basically the only component of spending that co-moves during that period in the three largest economies of the euro area (other spending components, or say unemployment exhibit a different, more idiosyncratic, behaviour). Fixed investment decreases after what could be labeled a large uncertainty shock (the increase in the btp/bund spread). The observation that fixed investment in UK increased (not shown) after the shock is an indication that the uncertainty concerns the euro. The Banking Union would signal a very strong commitment to the euro and potentially reverse the negative uncertainty shock. A positive certainty shock could even boost investment. The right panel of Figure 2 shows that the normalisation in the interbank market has not yet  been transmitted uniformly to the real economy across the three countries. The Banking Union should permit a more uniform transmission of the monetary policy allowing credit conditions to be more harmonised across the euro area.

Click to enlarge


Tuesday, 17 September 2013

Fiscal multipliers and Income share distribution

There is a literature that studies the consequences of fiscal consolidation on inequality. First a fiscal consolidation usually increases unemployment which is usually associated with a decrease in the wage share. The lower wage share increases inequality given the higher wage share in the total income of lower income groups. Second fiscal adjustments biased towards more regressive taxes (such as consumption taxes) or on cut-backs in progressive spending tend to increase inequality. Finally there is new evidence that fiscal consolidations success stories are related to taming inequality. Now there is a second issue which has not gained much attention in the empirical literature: is inequality related to the size of the multiplier?. This second question is natural if there is a mapping between credit constraint households and relatively low income households. If the link between inequality and the multiplier exists, fiscal consolidation policies that increase inequality are likely to have lower probability of success in their consolidation effort.
A recent influential paper by Blanchard and Leigh 2013 investigates the relation between growth forecast errors and planned fiscal consolidation during the recent financial crisis. They find that fiscal multipliers were substantially higher than those implicit in the models used by forecasters. Blanchard and Leigh also check if their results are robust to controlling of additional variables that could have possibly triggered both planned fiscal consolidation and lower-than-expected growth. The list of control variables is comprehensive and includes the initial debt ratio, the initial fiscal balance, the initial structural fiscal balance, the initial sovereign CDS, the initial bank CDS, a banking crisis dummy, the initial growth forecast, the initial potential growth forecast, the trading partner fiscal consolidation, the pre crisis current account balance, the pre crisis net foreign liabilities and the pre crisis household debt. Table 1 reproduces their baseline results for a sample of 26 european countries in 2011. The baseline coefficient is closed to -1 and is barely influenced by the inclusion of any of the controls. Remarkably none of the controls appears to be statistically significant. (click to enlarge)
Table 2 presents the same baseline regression augmented with a new set of controls that measure inequality. More precisely the controls are the income quartile, quintile, decile, fifth and first percentile share ratios. (click to enlarge)

They are calculated as the ratio of total income received by respectively the 25%,20%,10%,5%,1% of the population with the highest income to that received by the 25%,20%,10%,5%,2% of the population with the lowest income. All income measures are initial (end-2009) equivalized disposable income .(The equivalised disposable income is the total income of a household, after tax and other deductions, that is available for spending or saving, divided by the number of household members converted into equalised adults; household members are equalised or made equivalent by weighting each according to their age, using the so-called modified OECD equivalence scale.) and are shown in Figure 1. (click to enlarge)

 The coefficient on the consolidation forecast does not change, so that the results of Blanchard and Leigh still hold. However the coefficients on the controls are statistically significant and fairly large. Take for example the income quartile share ratio. The average income quartile ratio across the 26 countries in the sample is approximately equal to 4 implying that the 25% percent with higher income has on average 4 times the income of the 25% with lower income. The estimated coefficient on the income quartile share ratio implies that an increase from 4, Switzerland, to 5, Portugal, would lead to a domestic output loss of -0.93% larger relative to forecast in Portugal relative to Switzerland. This is of the same order of magnitude of the underestimation of the multiplier. Figure 2 collects the coefficients together with the 95% confidence intervals. (click to enlarge)

Recall that the controls are initial period value (end-2009) so that the causality is likely to go from distribution to growth forecast errors. Given that most of the countries in the sample have experienced a recession, the negative and significant coefficient on the income distribution measures is suggestive of standard Keynesian channels if the income distribution proxies for share of households that are credit constraints and/or have higher marginal propensity to consume. In any case further analysis must be performed to better interpret the results.

Thursday, 12 September 2013

Expanding times

The German economy is enlarging (the antonym of shrinking). Blue lines are flows, green lines are prices, red lines are stocks. Click figures to enlarge. I could not find the deficit/surplus data for the first and second quarter of 2013. I thought it could be interesting to know the public finance stance of the largest eurozone economy. However, der Teufel steckt in der Fussnote...
Certain transactions, including net lending (+)/ net borrowing (-) are under embargo until all quarters of the current year are available.

Mais aucun problème.
Certain transactions, including net lending (+)/ net borrowing (-) are under embargo until all quarters of the current year are available.