Monday, 4 January 2016

Boom, Doom, Gloom & Africa

"Africa´s Boom is Over", pronounced an article in Foreign Policy on New Year´s eve. One of the conclusions: "Without the commodities boom, the actual failure of Africa’s development has now been laid bare". The gloom echoes recent forecasts by the IMF, notably its Regional Economic Outlook: Sub-Saharan Africa 2015, released last October. An often-used conceit, just as in the IMF´s WEO released the same month, is that of "headwinds" that Africa encounters as raw material prices have crashed and large emerging economies are slowing down. Africa counts 10 net oil exporters and 15 net nonrenewable exporters among its 54 countries. While these 29 countries do not constitute a majority of Africa in terms of country numbers, they command a large majority of Africa´s GDP and population.

While the winds of change connected to the slowdown in the majority of emerging countries (with the notable exception of India) may well imply considerable headwinds, the rebalancing of China in particular may also provide backwinds for Africa. China´s reforms aim at rebalancing the composition of growth in China toward consumption and away from investment. Such reforms are compatible with a rise of the real exchange rate (higher prices for nontraded services relative to tradable manufactures), of inflation-adjusted wages and of the level of domestic absorption in China. Such a process may affect Africa in several ways:

·         The price of energy and industrial commodities drops as a result of both the slowdown and the rebalancing. The biggest winners are those countries with either large energy import needs or relatively fewer commodity exports, such as Kenya and Tanzania (where the fuel share of imports exceeds 25%), and to a lesser extent Ethiopia and Mozambique. Africa’s centre of economic gravity is thus likely to shift from west to east, to the less commodity-dependent economies of Ethiopia, Kenya, Mozambique, Tanzania, and Uganda. Investment finance will follow this shift, reinforced by the peripheral outreach of China´s One Belt One Road initiative that includes East Africa for infrastructure finance[i].
·         Prices for soft commodities should be supported by China´s rebalancing as coffee, tea and protein-based (soya, for example) are consumed and imported more than before. However, the supply elasticity of soft commodities is higher than for exhaustible resources, so the price impact should be contained. Still, higher export volumes (at stable prices) will translate into higher export proceeds and government revenue in African soft-commodity producers.
·         To the extent that rising wages in China lead to higher labour unit cost, external competitiveness in low-end manufactures will be eroded. With incentives for some industries to move offshore, part of this relocation will involve sub-Saharan Africa (such as to the East African  garment-production). China could expand its current presence in sub-Saharan Africa’s pilot special economic zones, or encourage creation of new ones. The relocation of Chinese firms into Africa should lead to increases in factor productivity and shifts in global trade shares from China to Africa. Thus, all of Africa might experience positive effects as countries in the region are able to build domestic industries based on China relocating a portion of its manufacturing base permanently to the region. 

To be sure, the relocation process takes time, and it takes longer for the gains to be realized than the immediate income losses suffered by commodity exporters. But it´s not all doom and gloom in Africa.

[i] China’s new Silk Road Fund is an acknowledgement of this. It seeks to facilitate trade links with a number of frontier and emerging markets through a USD 40 billion infrastructure investment fund. In Africa, the fund is targeting the economies along the eastern seaboard, which suggests a shift away from China’s traditional focus on securing natural resources towards one more focused on exploring the opportunities for establishing a manufacturing hub in the region.


Tuesday, 1 December 2015

FfD3 Hopes Meet EM Slowdown in Africa

Just a little while ago (last July), they were so hopeful when adopting the Addis Ababa Action Agenda of the Third International Conference on Financing for Development (FfD3):

“We, the Heads of State and Government and High Representatives, gathered in Addis Ababa from 13 to 16 July 2015, affirm our strong political commitment to address the challenge of financing and creating an enabling environment at all levels for sustainable development in the spirit of global partnership and solidarity. We reaffirm and build on the 2002 Monterrey Consensus and the 2008 Doha Declaration.”

… and so on…  The SDG tanker had been set on course, and little attention seems to have been  given to the slowdown in most large emerging economies (EM) nor to the risk that ample liquidity driven by OECD monetary policy will not last forever. Poor countries´ challenging economic backdrop dominated by falling commodity prices, lower demand from China and the prospects of the Fed eventually hiking rates, all that was largely ignored in the official FfD3 documents

But consider the prospects of mobilizing foreign and domestic resources in Africa, which has become increasingly China centric over the last 15 years. According to the IMF latest (October 2015) Regional Economic Outlook on Sub-Saharan Africa, aptly subtitled “Dealing with the Gathering Clouds”, Africa´s recent good macro performance has rested on three pillars: high commodity prices; associated capital inflows; and better policies and institutions. Wasn´t the Fund consulted during the FfD process? Or was it simply ignored?

Figure 1: Commodity Prices August 2015, 2016 Projections
% change since January 2013

Source: IMF (2015), “Dealing with the Gathering Clouds”, October.

The first pillar, commodity prices, has broken over the last three years (Figure 1). Sub-Saharan Africa counts 8 net oil exporters and 15 net nonrenewable exporters. The prices of their main export staple - fossil fuels and metals - have crashed by between more than 60 and 30 percent during that period. Oil exporters are hard hit: If oil exports constitute 40 percent of a country´s GDP, an annual drop in oil prices by 25 percent translates into an income effect of minus ten percent. To be sure, commodity prices may fall even further; they may stagnate from now on; and they may rise again. Only the rise of commodity prices would alleviate African producers. Don´t bet on it, says Carmen Reinhard (2015)[1]: price declines typically retain downward momentum. By the end of the boom, many commodity exporters had already initiated investment projects to expand production. As these investments bear fruit, the increased supply will sustain downward pressure on prices. For minerals, short-term supply is price-inelastic: a result of near-insurmountable barriers to exit and a significant proportion of fixed costs.

The second pillar, net capital inflows, is so far surprisingly unaffected prima facie. The IMF WEO, released last October, still projects a small surplus (0.4% of GDP) in the capital account for Sub-Saharan Africa. DB Research[2] produces fresh evidence that Eurobond issuance by Sub-Saharan African frontier sovereigns (excluding South Africa) has held up remarkably well in 2015.

Figure 2: Change in Debt Service Cost, Sub-Saharan Africa 2015

However, issuers had to offer significantly higher yields than previously and yields on secondary markets jumped to multi-year highs (Figure 2):
·         While Emerging Market Bond Index Global (EMBIG) spreads have moved up since October 2014 by less than 100 basis points (bp), sub-Saharan sovereign bond spreads have risen by between 120 bp (South Africa) and almost 500 bp in Zambia.
·         As all outstanding sovereign Eurobonds in SSA are denominated in USD, any depreciation will have a direct effect on the local currency value of debt service payments. This is potentially even more harmful for interest burdens than rising spreads. The drop since summer 2014 against the USD has been most severe for the Zambian kwacha, the Angolan kwanza, the Namibian dollar, the Ugandan shilling and the Tanzanian shilling which lost between 51% and 20% yoy against the USD, as of November 2015. Zambia has been hit hardest: debt service cost rose in 2015 by 18 percentage points of GDP (left hand scale in Fig. 2) and by 106% in local currency terms (right hand scale).

The difference between rising local currency cost of Eurobonds – around 20% on average in SSA - and the drop in growth rates to low single digits makes for terrible debt dynamics, which is unlikely to be compensated by a corresponding surplus in the non-interest account.[3]

This leaves us with the third pillar, better policies. In the realm of development finance, this translates above all into improved tax collection. Recent IMF analysis (“Dealing with the Gathering Clouds”) suggests that the median country in sub-Saharan Africa might have a potential for another 3 to 6.5 percentage points increase in tax revenue. AID spel

[1] Reinhard, Carmen (2015), “The Commodity Roller Coaster”, Project Syndicate, 9 November.
[2] Masetti, Oliver (2015), „African Eurobonds: Will the Boom Continue?“, Deutsche Bank Research, Research Briefing, 16th November, Frankfurt/Main.
[3] Debt dynamics for a country´s local currency debt-GDP ratio eD/Y are driven by e(r-n)D + (G-T), with e the real exchange rate, r real interest cost, n real growth rate, D/Y the debt-GDP ratio and (G-T) the non-interest deficit. The same debt-dynamics equation can be applied to public finance.

Sunday, 8 November 2015

Africa´s Resource Gaps are Tightening

The structural slowing of potential output growth in emerging market economies that led to lower commodity prices has been simulated in the latest IMF WEO, released in October 2015 (Scenario Box 1, p 25). In particular, the marked decline in investment and growth in China—together with the generalized slowdown across emerging market economies—implies a sizable weakening of commodity prices, particularly those for metals, resulting in a weakening of the terms of trade for commodity exporters. Lower expected growth leads to lower investment. Africa´s resource mobilization might tighten via the various resource constraints for investment and output. The Bacha (1990) three-gap model has highlighted the foreign-exchange constraint, the private domestic-saving constraint and the fiscal constraint[1].

Countries at early stages of development (optimally) pursue an investment-based strategy, which relies on existing firms and managers to maximize investment.  Most of Africa is still at that early stage. High domestic savings and investment rates have underpinned latecomer development in the 19th century in Europe and in the 20th century in Asia. In Africa growth after 2000 tended to be higher in countries with higher investment shares in GDP, as discussed at length in AEO 2014),and investment tended to be higher in countries with higher national savings. 

Table 1: Variables to Determine Financing Needs
Structural, rigid short term
·         Import content of investment
·         Crowding-in coefficient 
·         Global interest rate
Policy, rigid short term
·         Private savings 
·         Remittances 
·         Net factor payments abroad 
·         Net capital inflows: of which
direct foreign and portfolio equity investment
·         Private investment 
·         Exports 
·         Imports
·         Exchange rate, in monetary union
Policy, manageable
·         Public investment 
·         Change in FX reserves 
·         Exchange rate, unless monetary union
·         Net capital inflows, of which
Loans, foreign bonds and aid inflows

The gap model provides a consistent list of variables that matter for resource mobilization in Africa. To be sure, only some of those variables can be influenced by policy, at least in the short term. Therefore, the gap equations also provide the basis – as was their historical motivation 50 years ago already – to quantify Africa´s public financing needs. Table 1 attempts to classify the variables into three categories, although the underlying distinctions may be fluid and somewhat arbitrary: structural (rigid short term); policy (rigid short term); and policy (manageable). 

Investment in Sub-Saharan Africa has traditionally been constrained by low domestic savings. Only in the ´golden decade´ of the 2000s the region recorded a saving rate that averaged almost a fifth of its combined output.  Since 2009, Africa´s saving rate has been declining, and the IMF projects for 2015 the domestic savings rate to drop even further (Table 2), to a meagre 15.4 percent of GDP. This compares to an average saving rate of 31.9 percent projected for the total of emerging and developing countries in 2015. Sub-Saharan Africa is the developing region with the world´s lowest saving rate. First and foremost, investment and future output are saving constrained.

 Table 2: Financial Balances 2001-08 and 2015p,
SS Africa and Total Emerging and Developing Countries (EMDC)
- percent of GDP -

As domestic investment is projected to remain sustained at more than 20 percent of GDP in the Fund projections but savings are depressed, the current account is pushed into deficits, exceeding five percent of combined GDP in Sub-Saharan Africa. The projected deficit on Africa´s current account is consequently considerable, mostly financed by running down official reserves. This begs the danger of currency attacks, with subsequent currency mismatches and balance-sheet recession.

[1] Edmar l. Bacha (1990), “A three-gap model of foreign transfers and the GDP growth rate in developing countries”, Journal of Development Economics, Vol. 32, Issue 2, April, pp. 279–296, doi:10.1016/0304-3878(90)90039-E. While it can be objected that the gap model is too structuralist, it seems relevant in the current African situation as several constraints can be taken as given for short-term analysis.

Monday, 28 September 2015

Headwinds ahead for Piketty´s r>g?

Fast rewind some 40 years back:  The initial opening of China and India to world markets really became felt from the 1980s – a ‘one-off’ event that integrated 2 billion people or 40% of global labour force in the global market economy. The opening to trade increased the share of workers with basic education in the world labour force and lowered the world average capital/labour ratio. The relative endowments of other countries were thus shifted in the opposite directions, which tended to move their comparative advantage away from labour-intensive manufacturing (Wood and Mayer, 2012[1]).
The impact on real wages in advanced countries is easily captured in a simple Cobb-Douglas production function. With factor shares a third each for labour, capital, and Know How (Mankiw, Romer & Weil, 1992[2]), I had estimated that the mechanical opening impact on global real equilibrium wages was round 16.5% (Wolf, 2006[3]). A doubling of the global labour force with basic skills had halved labour productivity on impact; multiplied with the labour share of 0.33 produced the result[4]. Moreover, a sharp increase in the prime working age population added to the larger global labour force. The joint effect of the initial opening of the Asian giants and demographic factors pushed real wages lower and inequality much higher in the advanced economies.
Meanwhile, the drop in capital per labour combined with global imbalances supported corporate profits, capital returns and interest rates. Francis and Veronica Warnock had shown that international capital flows have an economically important effect on the most important price in the largest economy in the world, that of the ten-year U.S. Treasury bond[5]. Their analysis indicated that roughly two-thirds of the impact comes directly from East Asian sources. In addition, some of the foreign flows were owed to the recycling of metal- and petrodollars, as oil and metal exporters benefited from China´s motorization, urbanization and industrialization.
In his celebrated book[6], Thomas Piketty had established that capitalism had a fundamental force for divergence and greater wealth inequality, summed up in the inequality r>g. The formula relates the rate of return on capital (r) to the rate of economic growth (g), where r includes profits, dividends, interest, rents and other income from capital; g is measured in income (wages) or output. Note that Piketty in his book had conceded that the inert trend towards higher inequality was reversed between 1930 and 1975, due to some rather ´unique´ circumstances. Interestingly, 1975 coincides with the beginning of Shifting Wealth Phase I, which has come to an end a couple of years ago. Maybe, those circumstances that had disturbed that inert capitalist trend toward higher inequality were not that ´unique´, after all?
A new fascinating study headed by Charles Goodhart[7] marshal evidence to answer the question “Is Piketty history? We think so”.  The study focuses on the projected trajectory of global working age population (see Figure). Just as a larger labour pool pushed real wages lower and inequality up in the advanced countries, it is argued, a smaller labour force will lead to rising wages, a larger share of income for labour and a decline in inequality. The yearly rise in global working age population growth has peaked around 2005 at 70 million people; the rise is projected to drop to 30 million by 2040. China will actually face a shrinking labour force pool very soon, while Africa and India continue to see a rising labour force. Migration to advanced countries can dampen the positive wage effect but it must be massive (as in Germany currently).

Working Age Population, 1950 – 2040
yearly changes in million

Source: Morgan Stanley Research, based on UN Population Database

While the depressive impulse on wages is likely to attenuate as a result of changing labour force dynamics, the Morgan Stanley study foresees also an ageing-driven drop in capital returns. The advent of an ageing society will lead to a greater proportionate fall in personal saving than in personal sector investment (housing). The corporate sector is predicted to respond by raising the K/L ratio, i.e., by adding capital to compensate for the factor of production that is getting scarcer and more expensive. The overall rise in the K/L ratio, as the growth of the working population falls, is consistent with some decline in capital returns. However, interest rates are projected to rise as ageing lowers ex ante saving. As a result, the Piketty formula r>g might be replaced by w>r, with wages rise exceeding capital returns and inequality trends abating in advanced countries. Now that would run against against new conventional wisdom! 

[1] Wood, Adrian and Jörg Mayer, “Has China de-industrialized other developing countries?”, Review of World Economics, Vol. 147, 325 – 350.
[2] Mankiw, N.G., D. Romer and D.W. Weil (1992), “A Contribution to the Empirics of Growth”, Quarterly Journal of Economics, Vol. 107.2, May, pp. 407 - 437.
[3] Wolf, Martin (2006), „Answer to Asia´s rise is not to retreat“, Financial Times, 14 March. Martin cites slides of my Basel University lectures.
[4] δw* = δY/L = 1/3(-0,5) = 0,165, derived from Y = αL (1)K;  with δK/L = -0,5; L/Y = 1/3.
[5] Warnock, F. and V. Warnock (2006), “International Capital Flows and U.S. Interest Rates”, NBER Working Paper No. 12560, October.
[6] Piketty, Thomas (2013), Le capital au XXIe siècle, Paris: éditions du Seuil, August.
[7] Goodhart, C., M. Pradhan and P. Pardeshi (2015), Could Demographics Reverse Three Multi-Decade Trends?, Morgan Stanley Research, Global Issues, 15 September. Thanks to Prof. Goodhart for providing me with a copy.

Tuesday, 11 August 2015

Yuan Devaluation Signals Current Slump and Future Growth

On 11 August 2015, the People Bank of China weakened the yuan by almost 2 percent. The central bank’s move pushed the yuan´s “daily fix” to 6.2298 against the dollar, and that move was its biggest one-day change since China had loosened (somewhat) its dollar peg in June 2010. Financial market observers immediately jumped to two ´conclusions´:
·         While China´s central bank paid lip service to a more flexible exchange rate regime, it really signaled panic about China´s slowing growth;
·         The devaluation wasn´t a one-time shot but the likely entry to a prolonged period of engineering the yuan down to increase external competitiveness.
Consequently, all China-sensitive asset prices – such as oil, copper, shares of luxury brands and cars and emerging-market currencies – dived below their already depressed levels upon the devaluation news.
I will make the point that the market reaction and commentary is myopic: Even if the devaluation signals concerns about current growth, correcting the yuan´s accumulated overvaluation will be good for China´s growth, emerging countries´ growth as well as commodity and luxury exports.

Table 1: Yuan BIS Effective Exchange Rate, 01/2010 – 06/2015
2010 = 100

Some may still doubt that the yuan is overvalued. The evidence points to the contrary, though. According to the broad measure of BIS effective (trade weighted) exchange rate, the yuan has appreciated by some 30% over the past five years (Table 1). According to the Barclays behavioural equilibrium exchange rate model, the yuan remains the second-most overvalued currency in the world, by more than 20%[1]. The dollar’s recent ascent has pulled the yuan away from other currencies, leaving it increasingly overvalued, as the yen has stayed on a soft peg to the dollar. And even the Peterson Institute, traditionally the US Treasury´s ventriloquist to accuse China of unfair exchange-rate protection, has recently calmed down by gauging China´s so-called fundamental equilibrium exchange rate as ´correctly´ valued[2], at round 6 yuan per dollar. A further indication of yuan overvaluation is that the market and Chinese corporates had been expecting a yuan depreciation going forward: Onshore FX deposits surged in 2014, largely due to corporates holding more of their export proceeds in foreign currencies. Outflows recorded under the currency-and-deposits component of the balance of payments had also picked up, and forwards had been pricing in sizeable depreciation of the yuan versus the dollar.
Applying Dani Rodrik´s (2010) estimates[3] that a 10% nominal effective appreciation of the yuan would bring down China´s annual per capita growth by 0.86%, the 30% appreciation accumulated over the past five years may have chipped away some, 2.5% from China´s growth rate. This reduction in China’s growth has translated into a drop of GDP growth in poor countries (based on our former growth sensitivity estimates of 0.34 per cent)[4] by one percentage point of annual per capita income growth. The growth link between China and the emerging markets was even higher, with one percentage point of growth translating into 0.66% growth. These macroeconomic mechanics of growth can constitute a certain degree of growth optimism for China, the emerging countries, commodity prices, and world trade if China´s two percent devaluation in August is followed upon by the People´s Bank of China.

[2] William R. Cline (2014), Estimates of Fundamental Equilibrium Exchange Rates, November 2014, Policy Brief PB 14-25, November.
[3] Dani Rodrik (2010), ‘Making Room for China in the World Economy’, American Economic Review, Papers and Proceedings. Available at: aer.100.2.89
[4] Garroway et al. (2012), ´The Renminbi and Poor Country Growth, The World Economy,
doi: 10.1111/j.1467-9701.2011.01408.x

Thursday, 30 July 2015

China Growth Proxies: Aussie, Copper and VW

China´s nominal GDP growth has declined from a post-crisis (2009) level of close to 20% yoy to the recent low of 5.8% yoy in early 2015, according to the last CICC Macro Thematic Report. The government targets almost 10% nominal GDP growth, with 7% real annual growth. By contrast, a soft purchasing manager index (PMI) and another equity sell-off have reinforced fears of Chinese growth slowing down. The flash Caixin/Markit PMI released last week showed China's factory activity contracted by the most in 15 months in July; however, the official PMI stood at 50.2 in July, on par with the previous month's reading, according to the median forecast of 20 economists in Reuters poll.

Figure 1: China Growth, yoy, 2006-2015

What have the Australian dollar, copper futures and Volkswagen share prices in common? China. China´s linkage with the world economy is most directly read via demand-driven industrial commodities (copper, iron ore), market-determined China dependent currencies (such as the Australian dollar) and industrial goods that the Chinese like to and can increasingly afford (such as VW´s Audi cars). These asset prices can be taken as proxies for China´s growth.
China these days demands more industrial metals than the rest of the world combined. China´s industrialization and urbanization had created a super cycle in metal prices; with the new emphasis on consumption and services, this super cycle has probably ended. Still, short-term assessments about China´s growth prospects move copper futures and options contracts.
The Australian dollar (like the New Zealand dollar) is a flexible market determined currency, which has a high (ca 50%) beta to the renminbi (RMB). Several emerging-country currencies (which as a group had a 15-year low this week) will have even higher RMB betas; but their central banks intervene quite heavily on the FX markets.

Figure 2: Austral-Asian RMB Betas

Volkswagen depends on China for more than half its net profit and 71 percent of its free cash flow including income from joint ventures and royalties, according to auto industry analysts. Growing uncertainty over China has finally been weighing on Volkswagen shares. Half-year results show that weaker demand in the world’s biggest car market is already slowing the 89 bln euro giant.
If you believe (and many economists still do) that asset prices are leading indicators, a select narrow group of assets would provide information on China´s immediate growth prospects – the Australian dollar, copper futures, and, e.g., the VW share prices. If you think, instead, that China will defend its 7% growth target and soon start an upswing, you can consider the three assets as a great buying opportunity (and their prices as lagging indicators). Aussie, copper prices and VW shares are now caught in a nice solid downward channel.

Figure 3: Dr. Copper, 2011-2015

Figure 4: Australian/US Dollar, 2011-2015

Figure 5: Volkswagen Vz, 2015

Official Chinese macro data are routinely met with suspicion, especially by those foreign observers who think that government legitimacy is only supported by high growth and rising consumer wealth. Actually, sentiment about domestic economic conditions is nowhere in the world better than in China where 90% of the population consider the current economic condition as good, according to the Spring 2015 Global Attitudes survey released by the Pew Research Center. Although Pew Reseach doesn´t ask the Chinese about political (dis)satisfaction, the correlation between economic and political satisfaction is high; and political satisfaction nowhere higher than in Asia[1].
It would seem then that domestic assessments of China´s growth prospects are more upbeat than foreign (media, investors). To be sure, the foreign impact of China will morph as the People Bank of China is envisaging a wider trading band for the RMB, likely anticipating competitive depreciation as early as August. Historically, China has been very efficient to translate RMB depreciation into higher growth[2]

Time to go long Aussie, copper and VW?

[2] Rodrik, D (2008), “The Real Exchange Rate and Economic Growth”, Brookings Papers on Economic Activity, Vol. 2.

Monday, 22 June 2015

Addis and Multilateral Concessional Finance

The UN Conference on Financing for Development in Addis Abeba in July 2015 will pave the way for the implementation of the post-2015 development agenda. The GDI Briefing Paper series „Financing Global Development“ analyses key financial and non-financial means of implementation for the new Sustainable Development Goals (SDGs) and discusses building blocks of a new framework for development finance. Chris Garroway and Helmut Reisen will warn in their GDI Briefing Paper: Beware of "end poverty" euphoria and of trigger-happy reform of multilateral concessional finance.
The shrinking client base of concessional finance
The post-2015 agenda offers a transformative vision for ending extreme poverty by 2030 and shifting the world to a carbon-free growth path. Some of the institutions best equipped to finance this agenda, however, face pressure to downsize as borrower countries lose eligibility for concessional finance, as they surpass income per capita and creditworthiness eligibility criteria.
India, for example, is the largest World Bank borrower by volume, and recently became a middle-income country, with income per capita above the threshold for concessional finance eligibility. Other large, fast growing middle-income countries, including Vietnam and Nigeria, are already “blend” countries, receiving a mix of both highly concessional credits from the World Bank’s soft window, the International Development Association (IDA), as well as less concessional hard loans from the World Bank’s lending arm for credit-worthy countries.
The graduation from concessional finance of large middle-income countries raises a range of conceptual, political and operational issues for multilateral donors. On the conceptual side, the nature of global responsibility towards poor people living in non-poor countries – particularly those that have only just become non-poor – may conflict with political considerations against providing concessional finance to countries that have accumulated high reserves and are turning into emerging donor countries themselves. This raises questions about the future roles, mandates and instruments of IDA, but also African Development Bank (AfDB), Asian Development Bank (ADB) and their respective soft windows (which in the case of ADB is already in the process of being dissolved and merged with ADB's "ordinary capital resources"), as well as the International Monetary Fund (IMF) concessional instruments.
Prudence suggests approaching reform of the concessional windows with a precautionary, rather than deterministic, perspective to enable flexible institutional response. That perspective should not only consider future graduation prospects, but also prospects for reverse graduation, e.g., due to disasters, military conflict and governance failures.
Shareholders should not let multilateral soft windows blindly continue business as usual, but also must not allow them to ignore the option value of preserving their financial and institutional strength by “declaring success” and letting them shrink. 
The changing geography of poverty
The geographic distribution of people living in extreme poverty is changing due to the recent period of high growth. A number of the largest, faster growing countries continue to have sizable populations living in extreme poverty. In fact, three-quarters of the world’s extreme poor today live in countries classified as middle-income countries (MICs), which have limited access to soft finance. Only one-quarter of the world’s extreme poor live in the remaining 35 low-income countries (LICs).  Some believe that this will be a transitory phenomenon and that by 2025 poverty will become concentrated in fragile and conflict affected states. Others argue that a sizable share of world poverty could remain in stable MICs, or be concentrated in fragile MICs, e.g., such as Pakistan and Nigeria.
Many donors consider that middle-income country status itself is a reason to be reducing or even ending aid. The belief is that it is an affront to middle-income countries for donors to get too involved in their own internal policies, with the justification that distributional issues are fundamentally domestic in nature. A major political decision facing the international finance institution (IFI) shareholders is therefore whether the international aid community should target poor countries or poor people.
Medium-term poverty reduction scenarios published by the World Bank and others in recent years show that ending extreme poverty by 2030 is possible. The ambition required shouldn’t be underestimated, however. The release of the 2011 purchasing power parity (PPP) exchange rates estimated by the International Comparison Program also has prompted unsubstantiated claims that ending poverty may be easily accomplished, due to much lower price levels in developing countries than previously thought. This end-poverty euphoria and optimism is well intentioned, but misplaced. Jumping to conclusions about global poverty based on the new 2011 PPPs is unwarranted and potentially misleading. Leaving aside whether or not the new PPPs are appropriate for measuring poverty, using new PPPs will require re-calculating the poverty line itself to ensure it is the average of national poverty lines used in the poorest countries. The most thorough attempt to do this so far suggests a new global poverty line of $1.82 in 2011 PPPs, at which the world is still as poor as we thought, and ending poverty remains as ambitious (Prydz and Joliffe, 2015).
Regardless which PPPs are used, slow growth over the medium-term is the principal reason why ending poverty by 2030 remains highly ambitious. This is also why the high concentration of extreme poverty in middle-income countries is unlikely to be transitory. A study we conducted for BMZ (Garroway and Reisen, 2014) forecasts growth to 2025, measuring the impact on extreme poverty and on concessional finance graduation. Our main result: more than half a billion people will still live in extreme poverty in 2025, the majority in MICs ineligible for concessional finance.

Redefining eligibility criteria
At present, the multilateral soft-finance windows are all pegged to the IDA “operational cutoff”, which refers to eligibility ceiling defined by a specific level of GNI per capita (US$1,195 for FY13). This has been a malleable yardstick in the past; IDA eligibility was initially based on the “historical cutoff” (US$ 1,945 for FY13). High demand for limited IDA resources led shareholders to lower the cutoff in the early 1980s. Most observers agree that the cut-off itself is arbitrary and doesn’t reflect a salient distinction between countries at different levels of development.
There are possible avenues for redefining the present operational cutoff. IDA itself has suggested looking at other criteria beyond income per capita. Possible complements could be the UNDP Human Development Index, the UN Economic Vulnerability Index, or a similar index. Another approach would be to re-instate the original “historical” cutoff. The simple but elegant solution is not only justified by the historical precedent, but also finds support in the ultimate stated objective of concessional windows to provide a durable way out of dependence on external resources for development finance needs. Our BMZ study (Garroway and Reisen, 2014) point to the empirical fact that few countries with income per capita below $2000 have non-poor populations that can afford redistribution at the national level to cover the poverty gap. This capacity for redistribution – proposed by Ravallion (2010) – is measured by the marginal tax rate on the non-poor population necessary to close the poverty gap. It is an independent, highly relevant justification for raising the IDA operational cutoff back to its historical level.
Smoothening transition periods
The multilateral concessional windows can also soften the transition from soft financing by making “blend” status a more explicit step in graduation. Smoothing already happens in practice through the use of blended and hardened terms, but it is not part of an explicit phased and transparent approach. A widened transitional window for concessional resources could be made available to countries whose incomes fall between the current threshold and a higher threshold, perhaps the historical cutoff. Again this finds justification in the finding that roughly $2,000 per capita serves as dividing line between countries that can feasibly reduce poverty through redistribution and those that would face prohibitively high tax rates on the domestic non-poor.
IDA itself has proposed three criteria for access to transitional support: (a) GNI per capita below the historical threshold; (b) a significant poverty agenda, as measured by poverty levels and other social indicators; and (c) a significant prospective reduction in available financing after graduation. Such support could be made available for new graduates that meet these three criteria and would help smooth the transition of graduating countries. Given the eventual objectives of mobilizing domestic finance adequately, allocation under the transitional window might also be earmarked toward public spending for social inclusion and redistribution as well as improved fiscal federalism, i.e. through higher fiscal transfers from rich to poor states.
Strengthening sub-sovereign allocations
IFIs could also increase direct funding in grants or credits to local governments or even nongovernmental organizations in regions with per capita incomes below country-level eligibility thresholds, even if the country’s average income level is above the threshold. Apart from the rural-urban inequality in populous large emerging countries, such as Brazil, China, India or Indonesia, new threats like disaster-related impoverishment also have distinct within-country geography. Again, like the earmarking of transitional resources, the sub-sovereign allocations could also be aimed at cooperation on inclusive policy processes such as budget allocations, and sustainable urbanization to improve prospects for more inclusive development. Some MIC governments might interpret such cooperation as excessive interference into domestic political processes. But such concerns might be less for multilateral rather than bilateral donors. Within the multilaterals space this type of cooperation is well-suited for agencies where MICs have more voice in governance.
It also may be worthwhile to explore to what extent the European Union (EU) experience can inform multilateral concessional lending. EU Structural Funds and the Cohesion Fund are financial instruments of EU regional policy, intended to narrow disparities among regions and Member States. Since 2000, more than €500 billion in Structural Funds, mostly via the European Regional Development Fund, have been channeled to local projects in EU countries via national intermediary institutions.
Opening the soft windows for global public goods
Another major political decision relates to the role of multilateral concessional finance in provisioning global public goods, especially related to climate change adaptation and disaster management. An important side effect of mainstreaming climate change into development cooperation will be the need for multilateral donors to integrate vulnerability to environmental and global risks into their allocation criteria of concessional flows. An alternative to withdrawing concessional finance from MICs would be using the soft windows to co-finance regional and global public goods. The mandate of the soft windows could be adapted to focus explicitly on infrastructure with upfront cost but long-term developmental benefits as a way to help sustain global economic growth and human welfare. Tracer sectors could be climate change adaptation and disaster risk prevention and management.
Indeed eleven of the prospective 2025 IDA graduates we identified in our BMZ study already have greater than $US 100 million prospective annual disaster damage costs (Garroway and Reisen, 2014). Asian MICs, India and Bangladesh, top the list. Disaster risk management should be integrated with poverty eradication efforts; otherwise, ending poverty may not be within reach. The soft windows could thus contribute to the SDG-agenda while maintaining their anti-poverty mandate.
The option value of waiting on soft window reform
The option value of preserving the concessional windows is considerable in a world with global governance failures that prevent first-best policy solutions. The provision of global public goods requires the institutional infrastructure that these windows can deliver. Shrinking the soft finance windows prematurely would mean losing the considerable option value of waiting. It implies losing effective financial and technical services and know-how on a scale and with a quality that matter globally or regionally. It would also forego network externalities that represent a valuable global asset. Shrinking the multilateral soft windows would also imply – for better or for worse – that MICs would also need to speed up the establishment of new development banks, such as New Development (‘BRICS’) Bank and Asian Infrastructure Investment Bank, without benefiting from knowledge and ´certification value´ that existing concessional windows have acquired already (Reisen, 2015).

Garroway C. and Reisen, H. 2014. The Future of Multilateral Concessional Finance. Deutsche Gesellschaft fuer Internationale Zusammenarbeit (GIZ) GmbH.
Jolliffe, D. and Prydz, E., 2015. "Global poverty goals and prices : how purchasing power parity matters," Policy Research Working Paper Series 7256, The World Bank.
Ravallion, M. 2010. "Do Poorer Countries Have Less Capacity for Redistribution?," Journal of Globalization and Development, De Gruyter, vol. 1(2), pages 1-31, December.
Reisen, H. 2015. “Will the AIIB and the NDB Help Reform Multilateral Development Banking?”, Global Policy Journal, Wiley.