Perspectives 2019 2020 Public Sector

Citi Perspectives 15 2. De-risk using a correlated (local) currency. This can be accomplished by converting USD loans to more liquid currencies that are aligned with the specific frontier market to a greater extent. For example, currencies of major trading partners may have very close, substantiated correlation with the frontier currency, e.g. Mexican Peso and Colombian Peso, Kazakh Tenge (or Belarus Ruble) and Russian ruble, or Serbian Dinar and CEE currencies such as Czech Koruna, Polish Zloty and Romanian Leu (see figures 1 and 2). 3. Look beyond currencies to current account and FX drivers such as commodities. In some cases, a frontier currency does not correlate with the broader EM market or another liquid EM currency, such as that of a neighbor or trading partner. These countries typically have undeveloped money markets and restricted capital accounts. Examples include Ghana, Nigeria, and Ukraine. In these situations, borrowers could focus on the major drivers of country’s current account, which often impact their illiquid FX markets. For example, in Ghana there are three traded products that account for over 50% of exports — crude oil, gasoline and natural gas. If Ghana overlays a basket of its export commodities as a hedge for its USD obligations it would reduce USD/GHS risk by 70-80%. Similarly, Nigeria could use crude oil and natural gas to achieve similar FX risk reduction (figure 4). Ukraine could remove an estimated 70%-80% of its FX risk by linking its USD debt to contract prices for wheat or corn, which are its main exports. Over the medium term, using quarterly relative changes, Ukraine could substantially de-risk its USD borrowings. All three components (EM macro, currencies of trading partners and commodities impacting the current account) could be combined into a bespoke basket for a country to remove large parts of the FX risk that are beyond the control of government. The government would then be left with the residual FX risk, which is driven by local idiosyncrasies, politics and banking system issues. These risks are inherently unhedgeable by any proxy and probably the risk that the host country is most equipped to run. In short, frontier governments (and Development Banks) should insure risks that are beyond their control and are insurable, and retain risks that are most relevant to, and manageable by, them. RSD Spot CZK Spot Years 2008-2019 Figure 1: Serbian Dinar (RSD) Returns in Czech Koruna (CZK), 2008 — 2019 Source: Bloomberg and Citi KZT Spot RUB Spot Ave R = 94% Figure 2: Kazakh Tenge (KZT) Returns in Russian Ruble (RUB) 2015 — 2019 -20% -15% -10% -5% 0% 5% 10% 15% 20% GHS Returns Unhedged Hedged by Basket Mar-10 Mar-11 Mar-12 Mar-13 Mar-14 Mar-15 Mar-16 Mar-17 Mar-18 Figure 3: Ghana Cedi (GHS) vs Commodity Basket, 2010 — 2018 -30% -25% -20% -15% -10% -5% 0% 5% Unhedged Hedged by Basket Mar-10 Mar-11 Mar-12 Mar-13 Mar-14 Mar-15 Mar-16 Mar-17 Mar-18 Figure 4: Nigerian Naira (NGN) vs Commodity Basket, 2010 — 2018