The Global Financial Crisis (GFC) validated the buffer value of international reserves and the active management of buffer funds. These issues are especially pertinent for commodity exporters, for whom the high volatility of commodity terms of trade translates into large shocks that impact the real exchange rate and GDP.
The history of Latin America provides ample examples where adverse terms of trade shocks terminated spells of ‘good times’, leading to capital flight and financial crises. ‘This time has been different’ for countries that, opting for counter-cyclical macro policies during the 2000s, followed the dictum of ‘save for rainy day’. Chile has been a prime example of such policies (Céspedes and Velasco 2012, 2014). Frankel (2011) found that since 2000, fiscal policy in Chile has followed a structural budget rule that has succeeded in implementing countercyclical fiscal policy.[i] Furthermore, Frankel et al. (2011) found that over the last decade, about a third of the developing world has become countercyclical.
In Aizenman and Riera-Crichton (2014) we analyse the degree to which the growing importance of sovereign wealth funds (SWFs) – as well as the diffusion of inflation targeting and augmented Taylor rules – have impacted the post-crisis adjustment of Latin America to the challenges associated with terms of trade and financial shocks.[ii]
Our analysis focuses on the twelve largest Latin American economies.[iii] This set of emerging countries has the highest volatility in their commodity terms of trade (CTOT). The concept of CTOT follows Ricci et al. (2008), and differs from the traditional measure in that it only includes the relative prices of a country’s commodity exports and imports, weighted by their country-specific GDP shares. Another useful property of this measure of CTOT arises from the use of export/import over GDP as our weights; this allows us to reinterpret CTOT shocks as income shocks to the home economy and builds a direct link to effects on aggregate income and production. [iv]
Figure 1 plots the evolution of commodity terms of trade volatility over time, and the accumulation of international liquidity and composition of this liquidity over the last three decades. In spite of this high volatility, Latin American economies managed to achieve low output and real exchange rate (REER) volatility during recent decades, including during the Great Recession (see Table 1).
Table 1. Real output growth and real exchange rate in LAC 7
Source: Annual data was taken from IADB Macro Watch. LAC-7 includes Argentina, Brazil, Chile, Colombia, Mexico, Peru and Venezuela.
Table 2. Commodity-based stabilisation SWF in LATAM
Figure 1a shows how Latin American economies had started to accumulate large amounts of foreign reserves well before the last crisis, moving from an average stock of reserves of 7.5% of GDP before the Great Moderation to more than 15% after the Great Recession. Figure 1b showcases the appearance of SWFs as an alternative source of International liquidity. Most of these SWFs acted as automatic stabilisers following some fiscal rule dedicated to manage the windfalls from abnormally high prices of the commodities typically exported by each country.
Figure 1a. CTOT shock volatility vs. accumulation of foreign reserve assets
Figure 1b. CTOT shock volatility vs. accumulation of assets in stabilisation SWF
Figure 1C shows the country composition of the aggregate balance of stabilisation funds in the region.[v] Another relevant policy change adopted during the nineties was a set of explicit inflation targets – up to half of the countries in our sample became inflation targeters (see Table 3). Inflation targeting may represent an important part of our buffer story since inflation-targeting (IT) countries may decide to divert resources from REER stabilisation to internal price stabilisation.
Figure 1c. Figure 1C: SWF balances by country
Table 3. Inflation targets in Latin America
To gain further insight about these trends we look at 1980-2013, paying special attention to the period of relatively low volatility before the great recession (dubbed the end of the great moderation, 2002-2007), the great recession (2008-2009), and the post-recession period (2010-2013). We use a fixed effects error correction model to capture the effects of CTOT shocks and the dynamic adjustment of REER and Output Growth.
To investigate the potential differences in our buffer story over different sample periods, we use another layer of non-linearities to account for different economic structures such as degree of trade openness, indebtedness, exchange rate and IT regimes, and the like.
We confirm that both the stock of reserves and their active management reduce the effects of transitory Commodity Terms of Trade shocks to real exchange rate in Latin American economies. This ‘buffer effect’ seems stronger against risk of real appreciation than against risk of depreciation under relatively high levels of external debt, and in economies that are less open to trade. Fixed exchange regimes act as a substitute to reserve accumulation. Accumulation (de-accumulation) of reserves helps in buffering the transmission of positive (negative) CTOT shocks to output.
SWFs add new a dimension to foreign asset management. In contrast to reserves, SWFs seem important to buffer the REER from CTOT shocks with fixed exchange rate regimes and in relatively closed economies. SWFs also reinforce the effects of CTOT shocks on real output during negative shocks with fixed exchange rate regimes, and buffer the effect for relatively high external debt levels. Our buffer story seems strongest during the 80’s, 90’s, and the end of the Great Moderation (2003-2007). Yet during the great recession (2008-2009), we observe a disconnect between CTOT and REER, and the role of reserves. The REER-CTOT relationship seems to resume during the post-great recession period (2010-2013), and reserve buffering returns – but not at the levels observed previous to the crisis. The same story applies for active use of reserves, except that our buffer story returns in a stronger fashion during the post-recession period.
There seems to be a ‘substitution’ between reserves and SWFs, where SWFs take over the buffering of the REER and the real GDP during the Great Recession and the post-Great Recession period. Inflation targeting policy matters, potentially diverting resources to the preservation of domestic price stability – IT countries seem to give up the use of reserves to buffer against CTOT shocks, relegating this role to the SWFs. Monetary authorities in Latin American countries that appear to follow augmented Taylor rules place large weights on output gaps. Meanwhile, inflation seems to gain importance for IT countries. The nature of the regime matters – non-IT countries seem to switch from REER stabilisation targeting to an inflation target when committing to a formal IT rule. This potentially means that liquidity management is no longer used towards the stabilisation of REER under an IT rule. Instead, SWFs provide IT countries with an alternative form of liquidity management against foreign shocks when international reserves are committed to other macroeconomic goals. This is true for both REER and output growth stabilisation.
Our paper documents and validates the growing importance of liquidity management for commodity exporting countries. These policies mitigate the transmission of terms of trade shocks to the real exchange rate, thereby stabilising the domestic economy. We find evidence that SWFs may provide another margin of stabilisation. This role may be of greater relevance for IT countries, and in periods of heightened volatility. This division of labor is consistent with Tinbergen rule in policy design: to reach n targets, one may use n independent instruments. International reserves are useful in dealing with balance sheet exposure, and aiming at short and intermediate run stabilisation objectives. Yet hoarding international reserves is not a panacea, as the opportunity cost of reserves imposes a fiscal cost, and hoarding reserves may require sterilisation to mitigate their inflationary consequences. An inflation targeting regime may thereby relegate the goal of real exchange rate stabilisation to its sovereign wealth fund. Such a fund may have greater risk tolerance, and its accumulation impacts directly the fiscal stance and the real exchange rate. Remarkably, the buffering roles of reserves and SWFs do not need East Asian levels of hoarding – they are operative in Latin America at relatively modest levels of reserves/GDP and SWF/GDP.
While we focus on Latin America, the gains of combining IT with a fiscal counter-cyclical rule may apply to most countries with functioning institutions, beyond commodity countries managing SWFs. The logic of linking a country’s buffer funds (international reserves, SWFs, public debt) with the fiscal budget can be applied to any country once we recognize that the public debt is akin to Sovereign Liability Fund – a negative SWF. Similarly, by stabilising the economy in the presence of volatile CTOT and other shocks, counter-cyclical policy is a worthy goal for all countries. While properly managed IT deals with inflation and price stability, generically IT and Taylor rules do not suffice to stabilise both output and inflation (as has been vividly illustrated in the aftermath of the GFC and the ongoing Eurozone stagnation). The logic of Tinbergen rules of policy design imply that at times of large and persistent shocks, IT rules should be complemented with fiscal rules, as has been clearly illustrated by Chile and several other countries.
Aizenman, J and D Riera-Crichton (2014), “Liquidity and Foreign Asset Management Challenges for Latin American Countries,” NBER WP 20646.
Aizenman, J and R Glick (2009), "Sovereign Wealth Funds: Stylized Facts about their Determinants and Governance," International Finance, vol. 12(3), pages 351-386, December.
Aizenman, J and R Glick (2010), “Asset Class Diversification and Delegation of Responsibilities between a Central Bank and Sovereign Wealth Fund", NBER Working paper 16392, forthcoming, International Journal of Central Banking.
Aizenman J, M Hutchison, and I Noy (2011), “Inflation Targeting and Real Exchange Rates in Emerging Markets," World Development, 39:5, pp. 712-724.
Céspedes, L F and A Velasco (2012), “Macroeconomic Performance During Commodity Price Booms and Busts,” IMF Economic Review 60, pp 570–599.
Céspedes, L F and A Velasco (2014), "Was this time different?: Fiscal policy in commodity republics," Journal of Development Economics, Volume 106, January, Pages 92–106.
Céspedes, L F, R Chang, and A Velasco (2012), “Is Inflation Targeting Still On Target?” NBER Working Paper No. 18570.
Frankel, J (2011), "A Solution to Fiscal Procyclicality: the Structural Budget Institutions Pioneered by Chile," Journal Economía Chilena (The Chilean Economy), Central Bank of Chile, vol. 14 (2), pp 39-78, August.
Frankel, J, C A Vegh, and G Vuletin (2013), "On graduation from fiscal procyclicality," Journal of Development Economics, vol. 100(1), pp 32-47.
Mishkin, F S and K Schmidt-Hebbel (2007), “Does Inflation Targeting Make a Difference?” NBER Working Paper No. 12876.
Ricci, L A, G M Milesi-Ferretti, and J Lee (2008), Real exchange rates and fundamentals: A cross-country perspective, Vol. 8. International Monetary Fund.
[i] A crucial ingredient accounting for Chile’s success is that the official estimates of trend output and the 10-year price of copper – which are key to the decomposition of the budget in Chile into structural versus cyclical components – are made by independent expert panels and, thus, insulated from the political process. See also Céspedes and Velasco (2012).
[ii] See Aizenman and Glick (2010, 2014) for an overview of the diffusion of SWFs, and possible division of labor between SWFs and Central Banks. See Mishkin and Schmidt-Hebbel (2007), Aizenman, Hutchison and Noy (2011) and Céspedes, Chang, and Velasco (2012) for analysis on Inflation Targeting in practice.
[iii] Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, Mexico, Paraguay, Peru, Uruguay and Venezuela.
[iv] By construction, a percentage increase (decrease) in the commodity terms of trade measure is approximately equal to the aggregate net trade gain (loss) relative to GDP from changes in real individual commodity prices.See the Appendix of NBER working paper # 17692 for further details regarding the derivation of CTOT, Data Definitions and sources.
[v] While Chile has been the clear leader of the pack, accumulating close to 20 billion dollars in its copper fund before the crisis, Mexico, Colombia and later in the sample Peru have been increasing their SWFs thanks to windfalls from their energy (oil and gas) funds. Venezuela was able to accumulate a large amount of funds in the early 2000’s but then their fund was liquidated and has not been active since. See Table 2 for further details.