IMF 2018 Article IV Consultation for Greece

Recently, the IMF published its long-awaited Article IV consultation on Greece which includes an assessment of the latest developments of the Greek economy as well as its own DSA on Greek debt (which rests on significantly different assumptions than the ESM DSA).

The IMF starts with a stark chart showing how the Greek tragedy compares to the US Great Depression, the 1997 Asian crisis as well as the Eurozone crisis:

The depth of the Depression is quite similar to the US case while Greece has managed to «maintain» a 25% lower GDP for a period of 5 years.In contrast, even the US managed to return to its pre-crisis GDP level 7 years after the start of the Great Depression. This is a clear indication of the way the Greek case was tragically mismanaged by the European countries and the IMF whose priority was avoiding a principal haircut of official loans rather than a quick return of Greece to growth.

The IMF projects that Greece will grow moderately during the 2018-2022 5-year period which also coincides with the period during which the country will have to register primary surpluses of at least 3.5% GDP. Most of the growth is projected to come from fixed investment with private consumption contributing 0.5% annually and a neutral contribution from the foreign balance:

As I have outlined in the past, such a growth path rests on the assumption that Greek households will continue dis-saving at the order of €9bn annually even while they have already depleted their financial assets by €34bn in the 2011-2017 period. This is based on the fact that, given a neutral external balance and a 3.5% primary government surplus, sectoral balances indicate that the private sector will need to maintain a negative net asset position in order for the other sectors to achieve these balances.

Projecting nearly 1% annual increase in private consumption during the 2020 – 2023 period without any countervailing factor (such as a positive external balance or a significant relaxation in the fiscal stance) seems quite optimistic. An annual negative balance of just €8bn means that households will have to consume another €40bn of their financial assets in the 2018-2022 period. Only employment growth (which will increase disposable income of the household sector) will act as a countervailing force. It’s a pity that the IMF does not use sectoral balances to check whether assumptions for private consumption and government surpluses can be realistic in the long-run.

The other important part of the IMF document is obviously the Greek debt DSA as well as its assessment of the possibility of maintaining large primary surpluses for many decades.

In its baseline scenario the IMF staff agrees with the ESM that debt-to-GDP trends down and Gross Financing Needs (GFN) remain below 15% of GDP in the medium term.

Nevertheless, the IMF argues that Greece will be unable to maintain a primary surplus larger than 1.5% of GDP after 2022 while its long-run economic growth will hover around 1% in start difference with the ESM which is projecting a primary surplus of 2.2%. As a result, the IMF is much more pessimistic for the long-run, projecting that Greek debt will become unsustainable after 2040:

What is also quite interesting is how even medium-term sustainability rests on assumptions of large primary surpluses and growth during the 2018 – 2022. A small 2 year recession during 2019-22 (with a total of -3% GDP growth) coupled with a small primary deficit for just one year will immediately push debt-to-GDP close to 200% and GFN to 20%.

Lastly, the IMF staff try to justify analytically why Greece will be unable to maintain high primary surpluses and economic growth in the following years. While the specific arguments have been put forth many times in the past, it is interesting to repeat them here once more (in IMF exact wording):

• Ceteris paribus, aging would imply an average yearly decline of 1.1 percentage points in Greece’s labor force during the next four decades.
• Total factor productivity (TFP) growth over the last 47 years averaged just ¼ percent annually, by far the lowest in the Euro Area. Assuming this historical average TFP growth rate going forward, labor productivity (output per worker) would grow only at about 0.4 percent in the steady state (the rate of TFP growth adjusted for the labor share in output).
• Combining the historical growth in output per worker of 0.4 percent with expected growth in the number of workers of -1.1 percent would imply long-term annual growth of -0.7 percent.
• While studies have documented an impact on output levels of 3 to 13 percent over the initial decade, the impact of reforms on growth tends to fizzle out afterwards.
• Lifting long-term growth from its baseline of –0.7 percent to 1 percent requires reforms to add 1.7 percentage points to growth per year for the next decades. The OECD (2016) estimates that full implementation of a broad menu of structural reforms could raise Greece’s output by about 7.8 percent over a 10-year horizon, which translates into an increase in annual growth of some 0.8 percentage points for about a decade. Bourles et al. (2013) estimate this gain to be slightly higher, at about 0.9 percentage points per year, while Daude (2016) finds that reforms focused on product markets and improving the business environment in Greece could boost growth by about 1.3 percentage points per year for a decade.
• Implicitly, the 1 percent growth projection presumes that Greece would manage to increase labor force participation to levels that exceed the Euro Area average (to offset the significant projected decline in Greece’s working age population) and that would generate TFP growth rates permanently far above Greece’s historical average.
• Historically, Greece has been unable to sustain primary surpluses for prolonged periods. During 1945–2015, the average primary balance in Greece is a deficit of about 3 percent of GDP, although a brief period of near-zero primary balance took place at the time of Greece’s EU accession. The high water-mark for Greece was a primary surplus exceeding 1 percent of GDP during eight consecutive years (1994–2001).
• In a sample covering 90 countries during the period 1945–2015, there have been only 13 cases where a primary fiscal surplus above 1.5 percent of GDP could be reached and maintained for a period of ten or more consecutive years.
• Economic conditions matter. Among EU countries, before entering a period of high average primary balances, countries tend to have strong real GDP growth (2.7 percent) and modestly high inflation (4 percent). They also have moderate unemployment (10 percent) and low net foreign debt (24 percent of GDP), conditions that do not conform to those now applying in Greece. Moreover, during the high primary balance periods, growth has been rapid (about 3.4 percent), inflation slightly elevated (3 percent), and unemployment contained (at about 7.2 percent). This suggests that sustained periods of high primary surpluses are driven by strong economic growth rather than by sizeable fiscal consolidation.
• Unemployment weighs on the budget through higher social expenditures—such as for unemployment benefits and social safety nets—as well as lower income-related revenue. Greece’s unemployment rate is exceptionally high—only 10 countries have had unemployment higher than 20 percent in the postwar period.Within the above sample, the average primary balance corresponding to countries suffering double-digit unemployment rates is about zero percent of GDP (i.e. balance). For double digit unemployment lasting for 10 years or longer, the average primary balance is about -½ percent of GDP. With long-term unemployment likely to remain high for some time, pressures on social assistance spending in Greece—such as the guaranteed minimum income—are likely to mount.

Overall, the IMF tries its best to provide Europeans with political cover for the medium-term outlook on the Greek front while still presenting a scientific case for why the targets set in the Greek program are highly unrealistic and will not be achieved. In my view it should pay closer attention to sectoral balances which would make it even easier to argue why large primary surpluses cannot be maintained in a country with a structurally negative external balance.

The IMF on making Greek budget growth friendly

So the IMF decided to publish its views on the size and path of Greek budget surpluses. In a nutshell, it still thinks that only a 1.5% primary surplus target is credible in the long-run and even that target must be accompanied by «growth-enhancing» budget reforms, including a lower tax-free income threshold and a reduction in pensions in order to lower state transfers to the public pension system.

Yet if Europeans and the Greek government «agree» on a higher surplus target (the 3.5% target agreed by the recent Eurogroup meeting) then the latter has to legislate measures upfront in order to make that commitment credible.

My first comment is to state the obvious fact that there really doesn’t exist any sort of «agreement» between the Greek government and its European partners. Rather, European countries would like to avoid any actual debt relief and thus will demand higher surplus targets than those that are reasonable from an economic standpoint. It is quite obvious that the Greek government is the weak side of this bargain and, as long as the IMF thinks that any target above 1.5% does not make economic sense, it should pressure the Europeans (who are the strong side in this debate) into accepting deeper debt relief, instead of standing ready to work with any surplus target they demand from the Greek side.

The IMF cannot claim to be a neutral technocratic institution and yet sign-off budget balance targets which it clearly believes to be unrealistic from a technical point of view only because they seem to be the only ones «politically acceptable».

Turning to the specific details of the IMF analysis, there are a couple of points to be made.

The IMF believes that the tax-free income threshold is quite high in the Greek case, which results in more than half of the wage earners to be exempt from income taxes (while the Eurozone average is close to 8%):

Its proposal is to lower that threshold significantly in order to be able to reduce the «high marginal tax rates». My objections are two-fold: First, the IMF does not insist on such a reduction in order to strengthen the revenues of the social safety net and lower the tax rates of middle-income wage earners. Rather, it would like to see the additional revenue being used in reducing the tax rates of high-income earners (which stand at more than 50% if the solidarity tax is taken into account). Thus it is actually proposing a post-tax income redistribution from the low-income earners to the top. In the IMF view such a redistribution will be «growth-enhancing» although I personally fail to understand how its effects will be anything else but contractionary, at least in the short-term, since it will by definition redistribute income from persons with a low saving rate to individuals with a higher saving rate.

Moreover, Greece displays one of the highest «risk of poverty» rates in the Eurozone which is close to 36% (and is actually even higher for people 16-54 years old) as well as a significantly high Gini index. As a result, a reduction of the income threshold, especially if it is not used to strengthen the social safety net significantly, will result in a rise of the post-tax poverty rate and income inequality with ambiguous medium/long-term growth effects.

The second point of the IMF is that Greece makes budgetary transfers to the pension system that are many times higher than the rest of Europe, at 11% of GDP compared to 2¼ for the Eurozone.

Although it is difficult to deny that the Greek pension system is expensive, unequal and in need of reform, it is still true that the above analysis does not take the state of the economy into account. According to the latest Eurostat figures, Greece still posts an output gap of -10.5% compared to only -1% for the whole of the Euro area. As a result, a large part of the budget transfers to the pension system are not structural but cyclical, due to the high unemployment level (close to 25%) and the significant incidence of part-time, low-paying jobs for the individuals who are actually employed.

Based on the latest statistics for wage earners (March 2016), the part-time employment share is 29% (532 thousand persons) with an average salary of 405€ while the full-time average is 1220€ (1300 thousand persons). At the same time, the average old age monthly pension is close to 800€. It is obvious that no pension system would be able to survive without significant state transfers given the level of unemployment and under-employment present in the Greek economy.

One way to compare Greece with the rest of Europe in a cyclically-adjusted manner is to calculate old age pension expenditure as a percentage of potential product. This is exactly what I have done in the following table (nominal potential product is equal to potential output multiplied with the actual GDP deflator):

We can see that pension expenditure actually compares quite favourably with other European countries such as France, Italy and Portugal.It is thus probable that a large part of the state transfers are the result of the large economic slack present in the Greek economy. That suggests that Greece primarily needs cyclical relief (through lower surplus targets for instance) rather than an upfront deep structural reform.

The IMF Debt Sustainability Analysis for Greece

Yesterday the IMF released its debt sustainability analysis for Greece based on developments during 2015 (but not including the bank holiday and capital controls imposed after the referendum announcement). I consider it a very important document mainly because it shows (probably for the first time) how the basic assumptions of the adjustment programs were terribly optimistic and significantly disconnected from reality. It is also the first time that I know of that the IMF includes a (highly probable) scenario which requires Europe to write off part of Greek official debt (basically the Greek Loan Facility of €53bn) in order for the latter to become sustainable. It seems that the IMF has decided to catch up with reality. Having its largest ever program in arrears probably also played a role.

The first part of the analysis describes how some of the developments since June 2014 improved debt sustainability. These include:

1. Lower interest rates future path due to easing of monetary conditions in the Euro area which contribute a total reduction of €23.5bn in Greece’s debt up until 2022.
2. The return of the HFSF bank recapitalization buffer of €10.9bn. Obviously the IMF is not being honest in this point since the buffer will be needed in one form or another for further capital injections in Greek banks and/or for the creation of a bad bank to clear NPLs.
3. Intra-governmental borrowing of about €11bn to cover debt repayments which the IMF assumes that 2/3 will be sustained for the indefinite future by rolling over this short-term borrowing. This action will lead to an improvement of the debt-to-GDP ratio of 5% GDP.

On the other hand weaker GDP performance and downward revision of historical GDP contributed to an increase by 4% GDP of the 2022 debt-to-GDP ratio. Overall, taking all the above developments into account would improve the 2020 debt ratio to 116.5% (from a projection of 127.7% in the June 2014 review).

Yet at the same time the IMF has to acknowledge reality which includes much lower primary surplus targets, lower privatization proceeds, lower GDP growth, clearing arrears and rebuilding buffers as well as paying down a part of the short-term intergovernmental borrowing. This reality results in a total of financing needs of €52bn from October 2015 up to 2018 with the 12-month forward financing requirements from October 2015 amounting to €29bn.

Let’s take a closer look at a few aspects of ‘reality’.

First of all, primary surplus targets have been reduced from 3% for 2015 and 4.5% for 2016 onwards to 1% during 2015, 2% for 2016, 3% for 2017 and 3.5% for 2018 and beyond. Although in my opinion Greece should only be looking at the structural (cyclically adjusted) primary balance with official lenders taking the risk of short-term economic developments, these short-term targets are obviously much closer to the actual reality on the ground.

What is quite impressive is how the IMF has revised down its privatization proceeds targets. Projected privatizations were €23bn over the 2014-22 horizon yet only €3bn materialized during the last 5 years (the ‘fire sales’ argument of the current Greek government). As a result, the IMF now has much lower (and reasonable) targets of annual proceeds of around €500mn over the next few years. The magnitude of the targets revisions in each review is interesting:

What is the most important part of the document in my opinion is the analysis of long-term economic growth. The IMF acknowledges that its long-term growth target of 2% was unattainable and conditional on unreasonable assumptions and has now been updated to a still very ambitious target of 1.5%. It is clear from the document that even this target will most likely be missed. Only short-term targets based on closing the output gap and subject to a return of confidence are attainable in my opinion. In the words of the IMF itself:

Medium- to long-term growth projections in the program have been premised on full and decisive implementation of structural reforms that raises potential growth to 2 percent. Such growth rates stand in marked contrast to the historical record: real GDP growth since Greece joined the EU in 1981 has averaged 0.9 percent per year through multiple and full boom-bust cycles and TFP growth has averaged a mere 0.1 percent per year. To achieve TFP growth that is similar to what has been achieved in other euro area countries, implementation of structural reforms is therefore critical.

What would real GDP growth look like if TFP growth were to remain at the historical average rates since Greece joined the EU? Given the shrinking working-age population (as projected by Eurostat) and maintaining investment at its projected ratio of 19 percent of GDP from 2019 onwards (up from 11 percent currently), real GDP growth would be expected to average –0.6 percent per year in steady state. If labor force participation increased to the highest in the euro area, unemployment fell to German levels, and TFP growth reached the average in the euro area since 1980, real GDP growth would average 0.8 percent of GDP. Only if TFP growth were to reach Irish levels, that is, the best performer in the euro area, would real GDP growth average about 2 percent in steady state. With a weakening of the reform effort, it is implausible to argue for maintaining steady state growth of 2 percent. A slightly more modest, yet still ambitious, TFP growth assumption, with strong assumptions of employment growth, would argue for steady state growth of 1½ percent per year.

What the IMF also acknowledges is that any serious ‘return to markets’ from Greece will quickly make the current debt figures unsustainable because higher market rates will not be consistent with debt sustainability. As a result, in order for official creditors to avoid haircuts, more financing will be needed with concessional financing and a doubling of the grace and maturity period of loans.

It is unlikely that Greece will be able to close its financing gaps from the markets on terms consistent with debt sustainability. The central issue is that public debt cannot migrate back onto the balance sheet of the private sector at rates consistent with debt sustainability, until debt-to-GDP is much lower with correspondingly lower risk premia (see Figure 4i). Therefore, it is imperative for debt sustainability that the euro area member states provide additional resources of at least €36 billion on highly concessional terms (AAA interest rates, long maturities, and grace period) to fully cover the financing needs through end–2018, in the context of a third EU program (see also paragraph 10).
Even with concessional financing through 2018, debt would remain very high for decades and highly vulnerable to shocks. Assuming official (concessional) financing through end– 2018, the debt-to-GDP ratio is projected at about 150 percent in 2020, and close to 140 percent in 2022 (see Figure 4ii). Using the thresholds agreed in November 2012, a haircut that yields a reduction in debt of over 30 percent of GDP would be required to meet the November 2012 debt targets. With debt remaining very high, any further deterioration in growth rates or in the medium-term primary surplus relative to the revised baseline scenario discussed here would result in significant increases in debt and gross financing needs (see robustness tests in the next section below). This points to the high vulnerability of the debt dynamics.

In particular, if primary surpluses or growth were lowered as per the new policy package—primary surpluses of 3.5 percent of GDP, real GDP growth of 1½ percent in steady state, and more realistic privatization proceeds of about €½ billion annually—debt servicing would rise and debt/GDP would plateau at very high levels (see Figure 4i). For still lower primary surpluses or growth, debt servicing and debt/GDP rises unsustainably. The debt dynamics are unsustainable because as mentioned above, over time, costly market financing is replacing highly subsidized official sector financing, and the primary surpluses are insufficient to offset the difference. In other words, it is simply not reasonable to expect the large official sector held debt to migrate back onto the balance sheets of the private sector at rates consistent with debt sustainability.

Given the fragile debt dynamics, further concessions are necessary to restore debt sustainability. As an illustration, one option for recovering sustainability would be to extend the grace period to 20 years and the amortization period to 40 years on existing EU loans and to provide new official sector loans to cover financing needs falling due on similar terms at least through 2018. The scenario below considers this doubling of the grace and maturity periods of EU loans (except those for bank recap funds, which already have very long grace periods). In this scenario (see charts below), while the November 2012 debt/GDP targets would not be achievable, the gross financing needs would average 10 percent of GDP during 2015-2045, the level targeted at the time of the last review.

If grace periods and maturities on existing European loans are doubled and if new financing is provided for the next few years on similar concessional terms, debt can be deemed to be sustainable with high probability. Underpinning this assessment is the following: (i) more plausible assumptions—given persistent underperformance—than in the past reviews for the primary surplus targets, growth rates, privatization proceeds, and interest rates, all of which reduce the downside risk embedded in previous analyses. This still leads to gross financing needs under the baseline not only below 15 percent of GDP but at the same levels as at the last review; and (ii) delivery of debt relief that to date have been promised but are assumed to materialize in this analysis.

What is also important is that a reasonable scenario which includes of long-term growth close to 1% and a medium-term primary surplus target of 2.5% of GDP would require a haircut by official lenders:

However, lowering the primary surplus target even further in this lower growth environment would imply unsustainable debt dynamics. If the medium-term primary surplus target were to be reduced to 2½ percent of GDP, say because this is all that the Greek authorities could credibly commit to, then the debt-to-GDP trajectory would be unsustainable even with the 10-year concessional financing assumed in the previous scenario. Gross financing needs and debt-to-GDP would surge owing to the need to pay for the fiscal relaxation of 1 percent of GDP per year with new borrowing at market terms. Thus, any substantial deviation from the package of reforms under consideration—in the form of lower primary surpluses and weaker reforms—would require substantially more financing and debt relief (Figure 7).

In such a case, a haircut would be needed, along with extended concessional financing with fixed interest rates locked at current levels. A lower medium-term primary surplus of 2½ percent of GDP and lower real GDP growth of 1 percent per year would require not only concessional financing with fixed interest rates through 2020 to cover gaps as well as doubling of grace and maturities on existing debt but also a significant haircut of debt, for instance, full write-off of the stock outstanding in the GLF facility (€53.1 billion) or any other similar operation. The debt-to-GDP ratio would decline immediately, but “flattens” afterwards amid low economic growth and reduced primary surpluses. The stock and flow treatment, nevertheless, are able to bring the GFN-to-GDP trajectory back to safe ranges for the next three decades (Figure 8).

If one includes the fallout of the bank holiday it seems that we have clearly reached the end game. A recession during 2015 along with a low primary surplus will make the current DSA a bit outdated. Deciding to base the DSA on actually achievable targets (which in my opinion only include the ‘adverse scenario’ of 2.5% primary surplus target and 1% long-term GDP growth) will mean that Europe will (finally) have to consider a debt write-off. That is the economic reality on the ground. The way that the official sector reacts to it and the targets it (tries to) sets for the new Greek medium-term program will say a lot about who will bear the costs of further adjustment. In my view the policy of ‘externalizing’ the costs on the Greek economy and ‘extend and pretend’ is almost over, at least in democratic terms. Unfortunately, now will be the time of the politicians, name calling and trying to place the blame on the other party.

PS: It is also quite strange how the IMF puts such high value on ‘structural reforms’, when its own research shows that, at least ‘reforms’ that do not include more funding of investment in high-skilled labor, R&D, ICT capital and infrastructure but target the usual suspects of product and labor market have negative short-term and neutral long-term effects (labor market) or only marginally positive results (product markets). Only financing of serious investment in R&D, ICT capital and (to a lesser part) infrastructure can result in substantial effects on long-term TFP growth.

Zimbabwe hyperinflation

I recently came across a very interesting and thorough research paper from the IMF on the Zimbabwe hyper-inflation episode. What was quite clear from the document was that the high inflation of the period was the result of very high quasi-fiscal losses by the country’s central bank and not of runaway fiscal deficits. Probably the most important factor was interest costs of open market operations while also subsidies (mainly in the form of free forex to public enterprises) and foreign losses (due to a mismatch between foreign assets and liabilities) played a significant role. The annual flows were enormous, close to 60-80% of GDP which explains why annual inflation quickly reached levels of 600% and 1200%.

The rest of the post contains the relevant parts (and figures) of the paper that explain the hyper-inflation process more clearly:

While central bank losses in most countries have not exceeded 10 percent of GDP, Zimbabwe’s flow of realized central bank quasi-fiscal losses are estimated to have amounted to 75 percent of GDP in 2006. Losses have arisen from a range of activities including monetary operations to mop up liquidity; subsidized credit; foreign exchange losses through subsidized exchange rates for selected government purchases and multiple currency practices; and financial sector restructuring. Quasi-fiscal losses of this sort, rather than conventional monetary or fiscal laxity, have been the mainly responsible for the surge in money supply in Zimbabwe during 2005-7. The power to create money to finance losses quickly run into conflict with any recognized monetary policy objective with official inflation reaching 1,594 percent as of January 2007.

The following are noteworthy features of the balance sheet:

• Nonearning assets are substantial; they amounted to 83 percent of total assets as of October 31, 2006.
• RBZ securities, introduced as a sterilization tool at the beginning of 2004, became the largest liability by the end of that year, overtaking currency in circulation, previously the largest liability.
• Starting in 2004 sharp increases in statutory reserves to finance the concessional credit to favored sectors, such as agriculture, led to a steep climb in required reserves, which are not remunerated.
• Foreign liabilities, largely represented by credits from international financial institutions, have for some time been much larger than foreign assets. In this situation any currency depreciation produces losses.
• The smallest liability is capital and reserves which has been kept constant at a very low level. A central bank increases its capital through seigniorage and reduces it by operating expenses and distribution of profits to the government. Figure 1 shows the evolution of seigniorage since 2001 in Zimbabwe. For the RBZ, seigniorage has fallen from over 5 percent of GDP in 2001 to about 0.1 percent of GDP in 2005 because, given very high rates of inflation, real base money has declined drastically in relation to nominal GDP and the RBZ has invested in assets, including QFAs, with large negative real interest rates. Only the failure to apply a recognized accounting framework keeps the RBZ capital and reserve from being negative.

Most of the RBZ’s quasi-fiscal losses were incurred in connection with activities that go far beyond conventional central banking functions. There were four main sources of the losses:

• Subsidies in terms of free foreign exchange to public enterprises; price supports to exporters to partially compensate them for an overvalued exchange rate; and subsidized credit to troubled banks, farmers, and public enterprises.
• Realized exchange losses stemming mainly from the purchase of foreign exchange from exporters and the public at higher prices than sales of foreign exchange to importers (mainly government and public enterprises); and recognition of previously unrealized exchange losses upon repayment of external debt, including to the Fund.
• Interest payments associated with open market operations to mop up liquidity.
• Unrealized exchange losses reflecting official devaluations because foreign liabilities exceeded foreign assets.

To contain money growth, the RBZ sterilized the impact of the direct injection of liquidity into the economy that the QFAs represented. In January 2004 the RBZ started to issue its own bills at effective interest rates of over 900 percent per annum. These RBZ Financial treasury bills were naturally attractive to the market but too costly to the RBZ, so they were soon abandoned and replaced by Open Market Operation (OMO) bills, introduced in May 2004, and Special RBZ bills, introduced in June 2004. The OMO bills had the same interest rates as the existing government treasury bills but the accounting for them was clearly separated from holdings of government treasury bills since the interest cost was charged to the RBZ. The issuance of these bills escalated beginning in September 2004 after the large-scale financial or “liquidity” support to troubled commercial banks. The Special RBZ bills were introduced to absorb excess bank liquidity at the end of the day. They had a maturity of two years and carried an interest rate that was sharply negative in real terms. The long maturities deferred the monetizing consequences of the high nominal interest rates.

The RBZ has accumulated substantial domestic interest-bearing liabilities through open market operations to absorb liquidity. The vicious circle of rising losses and rising remunerated liabilities has resulted in inflation and increases in the interest rates of the bills, further accelerating the interest cost for the central bank. By 2005 the net interest cost of sterilization equaled 40 percent of GDP. In 2006 the interest cost grew further but its liquidity impact was partly alleviated as the authorities lengthened the maturity of treasury bills, thus deferring interest payments.

 

Greek Phillips Curve

The latest IMF WEO 2013 among other things includes an interesting chapter on recent developments on the link between inflation and unemployment. Compared to past recessions, deflationary pressures have been mild during the Great Recession:

These developments are attributed to the already low inflation (which creates a need for outright wage cuts in order to achieve lower inflation rates or even deflation) and anchoring of inflationary expectations due to monetary policy. Structurally high natural rates of unemployment do not seem to be present. One can then question whether the ‘divine coincidence’ still holds, where (based on the augmented Phillips curve) an inflation targeting central bank can maintain a zero output gap by achieving an inflation rate close to inflationary expectations. Since the Phillips curve appears to have become flatter, a low value for ε (in the APC π – π* = -ε [u -u*]) means that low inflation can coexist with large and persistent (as long as inflationary expectations are anchored) spells of unemployment. As a result, the current macroeconomic environment calls for a rethinking of central bank mandates closer to the Fed doctrine (which targets both unemployment and price stability).

This is quite evident in the case of Euro periphery countries such as Greece, where disinflation is accompanied with very large increases in unemployment. Since I ‘ve already touched upon the fact that the Phillips curve does not take into account firm profit margin changes, I will focus on the relationship between wage inflation and unemployment:

Starting in 2010 there’s a definite structural break in the rate of growth of ‘Nominal Compensation per Employee’ with wage increases turning negative and following the change in unemployment rate closely. During the 2006 – 2012 a simple linear trend is:

gw = 2.97 – 1.25Δu with an R²=0.86 while for 2010 – 2012 the trend becomes:

gw = 0.44 – 0.87Δu with an R²=0.90

It is clear that expectations of wage increases are close to zero (the 0.44 intercept) while the curve has become much flatter with the ‘unemployment multiplier’ falling below 1. That means that it will take more than 100bp changel in the unemployment rate to achieve a 1% fall in wages. Internal devaluation can only work through productivity from now on, especially since unemployment rates are moving closer to 30%.

Looking into the final demand deflator contributions one can observe the fact that most of the inflationary pressures since 2010 have been either transitory (import price and indirect tax effects) or the result of an effort by firms to maintain their profits margins (which given the fact that Greece is most probably going through a credit crunch are the only source of funding):

Labor and wages are the ones to have incurred most of the recession cost (with nominal ULC contributing a fall of -3.3% during 2012) while any remaining inflationary pressures seem to be the result of import prices and profit margins. The above suggest that a better targeted (and relaxed) monetary policy can be highly effective without fuelling any inflationary pressures since the Phillips curve is much flatter while a relaxation of credit constraints would allow firms to use bank credit again for working capital and investments instead of relying only on their own profits. That would allow them to adopt a pricing policy closer to domestic demand conditions and lower the CPI-based REER without hurting unemployment further.

IMF WEO 2013

The IMF recently released its World Economic Outlook for 2013 which among other things also includes some quite interesting research. I will focus on two issues, the divergence of politics between developed and emerging economies and the current account developments in the periphery countries

Divergence of Politics

Although world GDP has recovered since the Great Recession with the same pace as past recessions, this recovery masks different paths between developed and emerging economies with the former actually never achieving a reutrn to the output levels before the crisis:

Although monetary policy has actually been very accommodative during the latest downturn (compared with previous recessions), the same did not happen for real government expenditures. These were stable in developed countries and increased much faster than past recoveries in emerging countries. It seems that countries which adopted contractionary policies in terms of government expenditures actually experienced the heavier losses in output, mainly the Euro periphery and the UK.

Although correlation does not prove causation, the above data, coupled with the fact that monetary policy hit the zero level bound since the start of the crisis, point to the importance of fiscal policy to maintain aggregate demand and to large fiscal multipliers (with limited inflationary pressures).

Euro Periphery Current Account Developments

The IMF has done an excellent job of looking into the main drivers of pre-crisis current account imbalances and their post-crisis paths.  I ‘ve looked into Greek data in detail in the past and it seems that the WEO data point to the same conclusions. More specifically:

• CPI-based REER appreciation for Greece (but for other countries as well) was driven mainly by the NEER rather than CPI (or even ULC) differentials.
• Most countries were able to actually expand their export-to-GDP ratios driven by small ULC increases in the tradable sectors.
• What mostly happened was that ULC of the non-tradable sectors increased substantially in all countries (especially in Greece and Ireland) which led to imports growing faster than exports which had a negative effect on the trade balance.
• Overall current account deterioration (especially for Greece) was driven not only by the trade balance, but also from the income balance as well as transfers. Negative trade balances (absent exchange rate depreciation mechanisms) increased net liabilities with the RoW and led to negative income balances due to increasing interest payments.

In all countries, housing bubbles played a significant role since house prices increased much faster than the Euro area average and created a wealth effect which allowed domestic consumers to expand their borrowing and finance foreign goods purchases (while also enlarging the construction sectors in the relevant countries). In the case of Greece, the increase in ULC can be attributed to a large part in (almost double) increases in public sector wages which also drove private sector ULCs higher.

A few clear conclusions from the above analysis are that:

• Monetary policy could have played a large role in closing current account imbalances by lowering interest rates while maintaining the safe status of the securities held by the foreign sector. Lower interest rates would have allowed periphery countries to rollover their external liabilities at low rates and achieve a decline in both their fiscal and income deficits.
• A large part of the REER imbalances are driven by NEER forces which are basically controlled by the central bank rather than any individual country.
• The tradable sector did not seem to have actually lost significant competitiveness (as evidenced by the small increases in their relevant ULC) but rather the non-tradable sector grew larger and displayed high increases in wages. As a result, permanently closing current account imbalances will require costly and long-term deflationary policies with large lags while pushing unemployment, bankruptcies and permanent output losses to unacceptable highs in the meantime.
• A fiscal consolidation targeted at lowering public sector wage costs (through wage freezes and lower employment) in Greece seems logical since the public sector was the main source of the increase in ULCs.

Overall, the nature of the imbalances (large non-tradable sectors with substantial wage increases in the pre-crisis period) called for a stronger role of monetary policy (through lower rates, targeted asset purchases and exchange rate targeting) rather than extensive austerity since the rebalancing process would definitely take a lot of time and involve large domestic costs for the periphery countries.

IMF view on peak oil

Recently, IMF released a fine econometric study of oil production and prices. It examines both the peak oil view (which is mainly based on geology and projects a decline of global oil production in the near-term future) and the ‘economic’ view where price forces lead to substitution and increased production. It agrees with the geological view that, since 2003 geology has played a major role with global oil production maintaining a plateau without achieving the long-term historical growth rate of 1,5-2%. Coupled with a large increase of oil demand, especially from Chindia and a gradual drop in spare capacity oil prices sky-rocketed and had a serious role in pushing the world into the Great Recession.

It seems that the world oil production and economy has a ‘response function’ to high oil prices, mobilizing spare capacity and new oil fields (in the medium-term) and substituting away from oil. Nevertheless, the long-term outlook is one of a plateau in oil production and a doubling of real oil prices which could lead to serious economic shocks, especially to sectors where oil prices play a decisive role (air industry, tourism etc), although the IMF downplays the dangers, projecting world economic growth between 3-5%. Nevertheless they acknowledge the fact that the effects of high oil prices might become non linear and post serious threats to global economic development.

From a trading point of view, a long-term position on oil will probably provide for high returns, although they might include high volatility (since high oil prices will push the global economy to a series of recessions and a low growth/high inflation setup).

Below are a few key points from the paper itself:

Most importantly, the fact that the main output response to prices has been contemporaneous may be a reason for concern, because this indicates that output has mainly been able to respond to high prices by producers immediately dipping into spare capacity, rather than by increasing exploration or improving technology to increase longer-run capacity. To the extent that the future may be characterized by much tighter supply constraints and therefore much lower spare capacity, this option may no longer be available to the same extent as in the past.

We begin with Figure 7, the decomposition of oil prices. By 2008 oil prices had reached a

level that was 60% higher than what the model would have predicted on the basis of 2002 information. The major contributing factors in the earlier years were very strong oil demand, principally from booming emerging economies, and a positive world output gap. Oil supply, at least until some time in 2005, actually helped to, ceteris paribus, keep oil prices lower than what they would otherwise have been. From that time onward however, as we have seen, world oil production stayed on a plateau, and by 2008 insufficient world oil supply had become the major factor behind high oil prices. The Great Recession, from 2009, was so severe that oil prices dropped below the original 2002 forecast. The model attributes roughly half of this drop to a negative output gap shock, and the other half to a positive oil supply shock. The latter is the model’s interpretation of the increase in oil excess capacity in 2009. By 2011 real oil prices had regained their 2008 average (not peak) levels. The model attributes almost all of this to negative oil supply shocks, with oil demand and output gap shocks showing no major trend reversal after 2008. In other words, the insufficient growth of world oil supply that had begun to assert itself between 2005 and 2008 returned to center stage, as production remained on the same approximate plateau that it had reached in late 2005. It is very important, and evident from Figure 7, that it is not the shocks that are the major driving force behind the trend increase in oil prices in our model. Rather, the no-shocks scenario predicts an increase in oil prices that is not far from the actual trend. The reason is the significant estimate of the Hubbert linearization coefficient β1 in the oil supply curve. This confirms that the problem of oil becoming harder and harder to produce in sufficient quantities was an important factor that would have significantly increased oil prices regardless of shocks.

Our data and analysis suggest that there is at least a possibility that we may be at a turning point for world oil output and prices. A key concern going forward is that the relationship between higher oil prices and GDP may become nonlinear if oil prices become sufficiently high.

While our model is not as pessimistic as the pure geological view, which typically holds that binding resource constraints will lead world oil production onto an inexorable downward trend in the very near future, our prediction of small further increases in world oil production comes at the expense of a near doubling, permanently, of real oil prices over the coming decade. This is uncharted territory for the world economy, which has never experienced such prices for more than a few months. Our current model of the effect of such prices on GDP is based on historical data, and indicates perceptible but small and transitory output effects. But we suspect that there must be a pain barrier, a level of oil prices above which the effects on GDP becomes nonlinear, convex. We also suspect that the assumption that technology is independent of the availability of fossil fuels may be inappropriate, so that a lack of availability of oil may have aspects of a negative technology shock. In that case the macroeconomic effects of binding resource constraints could be much larger, more persistent, and they would extend well beyond the oil sector.

Update: An interesting overview of the paper in the Oil Drum

IMF fiscal austerity expansion view

I recently came across Chapter 3 of the IMF World Economic Outlook of 2010 which examines the macroeconomic effects of fiscal consolidation, the recent favorite European sport. It seems that even the IMF has already made clear how ineffective austerity is in the current environment, especially in a monetary union such as the Eurozone where exchange rate devaluation and monetary stimulus are off the table:

(abstract from the IMF analysis)

The main findings of the chapter are as follows:

• Fiscal consolidation typically has a contractionary effect on output. A fiscal consolidation equal to 1 percent of GDP typically reduces GDP by about 0.5 percent within two years and raises the unemployment rate by about 0.3 percentage point. Domestic demand—consumption and investment—falls by about 1 percent.

• Reductions in interest rates usually support output during episodes of fiscal consolidation. Central banks offset some of the contractionary pressures by cutting policy interest rates, and longer-term rates also typically decline, cushioning the impact on consumption and investment. For each 1 percent of GDP of fiscal consolidation, interest rates usually fall by about 20 basis points after two years. The model simulations also imply that, if interest rates are near zero, the effects of fiscal consolidation are more costly in terms of lost output.

• A decline in the real value of the domestic currency typically plays an important cushioning role by spurring net exports and is usually due to nominal depreciation or currency devaluation. For each 1 percent of GDP of fiscal consolidation, the value of the currency usually falls by about 1.1 percent, and the contribution of net exports to GDP rises by about 0.5 percentage point. Because not all countries can increase net exports at the same time, this finding implies that fiscal contraction is likely to be more painful when many countries adjust at the same time.

• Fiscal contraction that relies on spending cuts tends to have smaller contractionary effects than tax-based adjustments. This is partly because central banks usually provide substantially more stimulus following a spending-based contraction than following a tax-based contraction. Monetary stimulus is particularly weak following indirect tax hikes (such as the value-added tax, VAT) that raise prices.

• Fiscal retrenchment in countries that face a higher perceived sovereign default risk tends to be less contractionary. However, even among such high-risk countries, expansionary effects are unusual.

IMF paper on China’s future

I ‘ve already stated my views on China’s growth momentum and the main driving forces for it so far, based on various well written papers ([1], [2], [3]). I will summarize these views again here, in English this time:

China’s net export growth can be broken down in two distinct routes. A strong growth in exports and a stagnant increase in imports. The first factor can be attributed to:

• China’s membership in WTO increased its competitive advantage in labor-intensive sectors like textiles and furniture, strengthening its exports (up to a point at expense of other production countries).
• Metals and other related products (steel etc) took advantage of low energy costs (China is the number one user of coal in energy production in the world) as well as modernization of state controlled enterprises. The latter decreased their labor costs (they used to provide packages including housing, health care and pensions to their workers) and increased investments in expanding their production facilities.
• Machinery and technology sectors like LCD screens, laptops and cell phones gained in production, due to low labor costs, huge internal market and mass production capabilities hard to find in other countries in the world.

In general, the dollar peg was not the driving force for China’s growth but mainly allowed for both prices and FDI profits to have a low currency risk.

As for imports, their growth has not followed exports, creating an ever increasing trade surplus. The main driving force has been a very high savings rates (especially compared to its main export markets), which can be attributed to structural factors inside the Chinese society:

• Productive age population has been on the increase, with rural workers/farmers moving to urban centers pressuring wages. Furthermore, private house ownership (instead of state/company provided) is on the rise, pushing for a high savings rate with savings invested in common bank accounts.
• Especially since the 90’s, state enterprise employment was lowered, while the social security fabric was deregulated, both the health and pensions sector. Private means of protection gained weight, with workers saving in order to create a readily available safety liquidity buffer.
• The financial and private insurance sector is still far from mature which, coupled with strong capital controls (which place barriers in private investment abroad), create an insentive to invest savings in bank accounts, with commonly negative real interest rates.

My view has been that export growth will face barriers shortly, both due to energy costs (domestic coal usage has reached a peak, oil production has not much room to grow, China now functions as a ‘price setter’ for global commodity/energy prices) and because it now accounts for a large part of world exports, removing export share (and income) from other developed countries, although we might see China moving to higher quality product sectors (like cars for example). In the internal front, urbanization will face limits sooner or later, while the ‘one child policy’ will kick in, leveling the labor force numbers. The savings rate is an open question. Being more of a structural factor, it will probably play a key role in the medium to long-term, stabilizing the terms of trade.

IMF released an excellent, detailed paper on recent developments in China and its future perspectives. I will copy a small part of it here, though i strongly recommend reading all of it:

The sustained strength in imports of commodities and minerals has reinforced a dynamic that has been at work for several years now, going back to well before the global financial crisis. Over the past several years, imports have become more linked to commodities and minerals, where supply is relatively inelastic and global prices have been rising. At the same time, exports have become increasingly tilted toward machinery and equipment where supply is relatively elastic, competition is significant, and relative prices have been falling.

As a result, aside from 2009, China’s terms of trade has been steadily worsening. This may not be surprising from a historical context. Several other economies that have witnessed export-oriented growth (notably Japan and the NIEs) were affected by similar terms of trade declines along their development path. In China’s case, this dynamic is further fueled by the fact that, in both export and import markets, it has become so large as to no longer be a price taker. As a result, to some modest degree, China may well be generating a decline in its own terms of trade, creating a self-equilibrating mechanism that drives China’s terms of trade and creates countervailing downward pressures on China’s external surplus, through global prices.

Finally, a few words on the net income flows are warranted. It is something of a puzzle that, as China’s net foreign asset position has accelerated, there has not been a corresponding increase in net income flows. Looking first at the asset side, the rise in foreign assets in China has largely tracked the evolution of the central bank’s reserve portfolio and China Investment Corporation’s (CIC) investment position. Liabilities, on the other hand, have mirrored the growing stock of FDI flowing into China. The low net income flow numbers (which were actually negative in 2011) suggest a significant differential between the return on FDI into China versus that on the reserve holdings of the central bank. Indeed, it would appear, for much of the past several years, that return differential has been of the order of 3−4 percentage points, resulting in net income flows that are close to zero despite a growing net foreign asset position.

The Chinese government has rightly focused its policy efforts since the global financial crisis in a range of areas designed to accelerate the transformation of the Chinese economic model, improve livelihoods, and raise domestic consumption. Access to primary health care has been improved through the construction of new health facilities, particularly in previously underserviced rural communities. A new government health insurance program has been launched nation-wide, with the objective of achieving near universal coverage by the end of 2012 and subsidies for a core set of prescription drugs have been introduced. In addition, the existing government pension scheme is being expanded to cover urban unemployed workers across the country by end-2012 and to make those pensions more portable within China. Also, the absolute level of pensions has been increased, particularly for the elderly poor.

In addition to health care and pensions, improving access to affordable housing has been an important policy objective. The 12th Five Year Plan, launched in 2011, aims to construct 36 million low income housing units by 2016. The wider availability of low cost housing has the potential to ease the budget constraints of low income groups and release savings currently locked up toward financing home purchases.

Internal Devaluation – The IMF view

The latest IMF staff report on the Greek economy contains a detailed analysis of the fund’s view on the economic situation in Greece and its future prospects. Key elements of the report include:

• The economy is expected to contract by 5% in 2012, with the debt ratio leveling at 167% of GDP in 2013.
• Bank recapitalization due to the PSI and NPLs is projected to approach 50 billion €, lowering gains from the PSI deal (actually government debt will increase since new funding for the period 2012 – 14 will be close to 173 billion €).
• Budget adjustments of 2.75% of GDP are required for 2013 and 2014, in order to achieve a sustained primary balance of 4.5% of GDP (2012 balance projection is close to 1%).
• Potential growth is calculated around 2%, while long-term growth is lowered to 1-1.5%, making long-term debt sustainability challenging.
• ULC competitiveness deficit stands at 15% and is required to be eliminated within the next 3 years, mainly through real wage adjustments.

Overall, the task of stabilizing the Greek depression, without destroying the social fabric seems challenging (to say the least). Debt sustainability is based on almost impossible terms, which include strong continued fiscal consolidation of more than 5.5% of GDP,  closing of the competitiveness gap and a sustained rebound in economic activity within the following year.

What is very interesting is the fund’s own view on the capacity of internal devaluation schemes to produce working results, especially compared to the route of currency devaluation. In general, it seems that even the IMF is not sure about its own medicine having positive effects, while projected cumulative output loss is comparable to those of Argentina and Latvia.

IMF Internal Devaluation View

Internal devaluations are almost inevitably associated with deep and drawn-out recessions, because fixed exchange rate regimes put the brunt of the adjustment burden on growth, income, and employment. Depending on the size of the imbalances, the strength of adjustment measures, and the responsiveness of key macroeconomic variables, the duration of the initial adjustment period has ranged from 5 quarters (Hong Kong) to  15 quarters or more (Argentina before abandoning convertibility), while the depth of the downturn has varied from shallow growth recessions (Germany, Netherlands) to deep economic collapse accompanied by devastatingly high unemployment and emigration (Latvia).

Restoring competitiveness by way of internal devaluation has proved to be a difficult undertaking with very few successes. Countries with outright exchange rate devaluations usually recover faster.

Το σλάιντ απαιτεί την χρήση JavaScript.

Country experience suggests several factors are needed for internal devaluation to work. The most important preconditions are an open economy with high factor mobility and a high degree of wage and price flexibility.

Despite deep nominal declines in wages and pensions, real effective exchange rate depreciations have been regularly only modest due to only limited pass-through to prices (Baltic states, Argentina, Greece). Furthermore, private sector corporations are more likely to cut employment than to fully adjust wages, even in fairly flexible labor markets (Latvia). It also takes a long time for resources to shift from the non-tradable to the tradable sector, and both persistent skill mismatches and lack of increased investments in the tradable sector preclude full factor reallocation (former East Germany, Latvia). External adjustment therefore works predominantly through import compression rather than an expansion of exports—and oftentimes imports contract long before any real depreciation of the exchange rate. Finally, the often observed deterioration in asset quality and large increases in non-performing loans suggest that balance sheet effects are not limited to outright exchange rate devaluation—they only materialize more slowly in the process of internal devaluation as incomes fall but debt service does not.

The experience of Argentina in 1998–2002 shows that an economy can get trapped in a downward spiral in which adjustment through internal devaluation eventually proves impossible, and the only way to an eventual recovery remains default and the abandoning of the exchange rate peg.

Argentina ended convertibility in January 2002, almost four years into a deep recession that saw a 20 percent cumulative loss in output, culminating in sharp increases in interest rates, bankruptcies, unemployment, and poverty; deep cuts in wages and pensions; deteriorating asset quality, and deposit runs. The banking system collapsed and economic activity came to a virtual standstill in the first quarter of 2002. Nevertheless, only one quarter later the economy embarked on a rapid and sustained recovery, achieving 8.5 percent average real GDP growth over the following six years. The pre-recession output peak was exceeded after three years. Interestingly, despite a large and permanent real depreciation of more than 50 percent and a significant price boom in Argentina’s agricultural export products during this period, net exports contributed positively to GDP growth only in 2002, before turning negative again in the following years.