Enhancing the Monetary Policy Framework in Korea1
Inflation targeting from 1998 led to low and stable inflation, but since the global financial crisis Korea had faced more challenging conditions. As inflation expectations fell below the inflation target in 2012, the target was reduced by a percentage point to 2 percent in 2016. This weakened the credibility of the nominal anchor provided by the target, which monetary policy can help rebuild. This Selected Issues chapter outlines a strategy to facilitate this and navigate the more challenging monetary environment, involving enhanced communication of policy interest rate intentions and inflation-forecast targeting.
A. Experience with Inflation Targeting in Korea
1. The adoption of inflation targeting (IT) in 1998 contributed to low and more stable inflation in line with the target. This success partly reflected the benign global environment, associated with “the great moderation;” but, then, the more challenging environment after the global financial crisis (GFC) put inflation targeting to a sterner test. Inflation and inflation expectations fell below the bottom of the 2.5–3.5 percent inflation target range after 2012, as activity slowed. In 2016, the target range was replaced by a level inflation target of 2 percent, a reduction of a percentage point relative to the mid-point of the target range.
Korea - Actual and Expected Inflation
2. The reduction in the inflation target in 2016 created monetary policy challenges. The change was made as part of the Bank of Korea’s (BOK) regular triennial review of the monetary policy framework and aligns the target with that in most advanced economies. It was based on a reassessment of the optimal inflation target, which has declined owing to structural changes in the inflation process stemming from factors such as rapid population aging. While the adjustment of the target down for these reasons may be appropriate, the appearance that this was done to match the lower inflation rate (rather than ease aggressively to push inflation back to the higher target) could undermine monetary policy credibility. With inflation now close to the lower, revised target, the credibility of this new anchor should recover gradually over time. Monetary policy can accelerate the establishment of the new target as a fully credible nominal anchor by enhancing communication of how policy will achieve this target.
3. One measure of the effectiveness of inflation targeting is how well the path of the policy rate is reflected in forward interest rate curves. When this is the case, policy rate moves are passed through more fully into longer-term market interest rates, implying stronger monetary transmission. In the aftermath of the global financial crisis this appeared to be the case in Korea. The slope of these forward curves correctly anticipated that the policy rate would be raised, as reflected in the steep upward slope, following the very sharp rate cuts at the peak of the crisis. In contrast, in the second cycle of rate cuts starting in 2012, forward curves were downward sloping after the first rate cut; but, after that, reverted to an upward sloping or flat path, which shifted down with each policy rate cut. This suggests that after each rate cut, market participants were not expecting additional cuts, which tends to weaken the transmission of monetary policy.
Policy Rate and 2-Year Forward Rate Curves at Annual Intervals
B. Strengthening the Credibility of the Inflation Target
4. Inflation targeting became more challenging after the GFC. The larger shocks to inflation and output resulted in more persistent output gaps and deviations of inflation from target, making it harder to keep inflation expectations well anchored. Monetary conditions became harder to gauge owing to greater uncertainty about the size of the output gap and level of the equilibrium real interest rate, greater capital flows and currency volatility, and heightened risks to financial stability. This increased the scope for conflicts among policy objectives that can undermine monetary policy credibility by increasing incentives to deviate from the inflation target objective.
5. The more challenging environment revealed areas where the inflation targeting framework could be strengthened. The goal is to more firmly anchor inflation expectations to the target. This is necessary in an environment where inflation deviates from target for prolonged periods, as it is more difficult to verify if the central bank is acting to achieve its target. It requires that central banks convincingly explain how inflation is being brought back to target.
6. Enhanced communication of policy rate intentions can play a key role in strengthening credibility. BOK monetary policy decision making transparency is high with, for example, publication of MPC members’ views on economic conditions and how policy should respond, which are summarized in a statement by the Governor. Communication could be further enhanced through more forward-looking communication of how the inflation target is to be achieved. This involves characterizing the central bank “policy reaction function” by explaining how the BOK proposes to adjust policy rates in the future as circumstances change.
7. More effective communication can strengthen monetary transmission. As a short-term money market rate, the policy rate has little direct impact on economic activity. Rather, it works by moving longer-term rates—the rates at which households and firms borrow and lend—in the direction of the policy rate change. This requires that current changes in the policy rate affects expectations of the future path of the policy rate; which, in turn, influences the slope of the yield curve and, hence, longer-term rates. Key to achieving this is effective communication of monetary policy intentions so that markets will correctly anticipate the future policy rate changes.
8. Credibility can be strengthened by providing an inflation forecast and explaining how the policy rate will be adjusted so inflation converges to the target over the medium term. This approach of inflation forecast-targeting (IFT) does not require that the central bank systematically achieve its target, which is generally not feasible as large, unanticipated shocks hitting the economy cause realized future inflation to deviate from the forecast. Rather, credibility can be strengthened by communicating how the policy rate will be adjusted to return inflation to target and close the output gap over the medium term. This reveals how the central bank intends to navigate the short-run trade-off between inflation and output to achieve its target.
9. IFT is applied by central banks differently depending on the features of their financial system. Their communication of policy rate intentions to achieve the inflation target should evolve with experience and the sophistication with which markets and the public understand policy. It need not involve a numerical path for the policy rate; but, rather, a qualitative discussion of how the central bank is likely to respond to shocks impacting the inflation forecast. BOK communication already includes elements of IFT, which it can build upon. Evidence from the experience of central banks practicing IFT suggests that it strengthens the nominal anchor provided by the inflation target.2
C. Challenges Facing Korean Monetary Policy
10. Monetary policy challenges stem from the more uncertain environment. Four key sources of uncertainty are the size of the output gap, the level of the equilibrium real interest rate, capital flow and FX volatility, and the stability of the financial system, as outlined below. A risk-management approach can help policy navigate this more uncertain environment. It involves responding more strongly to larger shocks than to small ones, particularly large negative shocks where there is a risk that interest rates will hit the Effective Lower Bound (ELB). In this situation, more aggressive rate cuts lead to a sharper depreciation of the exchange rate and rise in expected inflation and, hence, a larger fall in the real interest rate, which magnifies the stimulatory effect.
The output gap. The output gap, a key gauge of inflationary pressure, has become more difficult to estimate owing to uncertainty about the extent of slowdown in potential growth. For example, a recent estimate is that potential growth had declined from a range of 3.0-3.4 percent during 2011-15, to 2.8-2.9 percent in 2016-20.3 When the output gap is thought to be small, high uncertainty means that its hard to know whether the gap is closed, negative, or positive. In this situation, a cautious approach can help protect monetary policy credibility from policy mistakes. This entails holding the policy rate steady for an extended period until incoming data or a new shock create a compelling case for a change. In contrast, when there is a large shock that widens the output gap significantly, the risk of policy mistakes owing to uncertainty about the gap is minimal and a forceful policy rate response would be appropriate.
The equilibrium real interest rate. The difficulty estimating the extent of decline in the equilibrium real interest rate—the rate that stabilizes the economy at full employment normalizes monetary conditions—is a source of uncertainty. When the policy rate is thought to be close to this level, there is a risk that monetary tightening could push the real interest rate above its equilibrium level, implying that the central bank needs to exercise more caution in raising policy rate.
High capital flow and currency volatility. Communication of the role of the exchange rate in monetary policy becomes more challenging in an environment of high currency volatility. When this is the case, explaining the key role that the flexible exchange rate plays in monetary policy transmission and as a shock absorber becomes more difficult. It requires a clearer, more explicit commitment to flexibility that involves resisting pressures to dampen FX volatility. This involves making a convincing case for a fully flexible exchange rate by clarifying how it helps insulate domestic monetary conditions from foreign monetary shocks by, for example, limiting the pass-through of U.S. rate hikes into domestic interest rates. This role is illustrated by the strong correlation of the won and the output gap over a relatively long time horizon (Figure). The communication challenge is that at short horizons this role of the exchange rate may be obscured by its high frequency volatility. Signaling a strong commitment to exchange rate flexibility strengthens credibility by reassuring markets that monetary policy will not be diverted from the inflation objective to dampen FX volatility.
Financial stability. When price stability and financial stability objectives come into conflict, communication plays an essential role in clarifying that policy rate decisions will be guided by the former rather than the latter. This challenge arises because the BOK is accountable for both objectives, which gives it a responsibility to clarify how it will achieve them when they come into conflict. For example, when the policy rate needs to be low in response to weak demand but the financial system is vulnerable, the central bank has incentive to raise rates to counter excessive risk taking, which can undermine monetary policy credibility. In this situation, the central bank needs to clarify how macroprudential policy can be used more effectively to target financial stability and that this allows monetary policy to focus exclusively on achieving price stability.
Estimated Declines in the Equilibrium Real Interest Rate in Korea Compared to Other Economies
Sources: IMF, International Financial Statistics; Haver Analytics; and IMF staff estimates.
Korean Output Gap and Won Exchange Rate
Source: Output gap: authors’ estimates. Exchange Rate: Haver Analytics.
D. Integrating Model-based Policy Analysis into Monetary Policy Decisions
11. Inflation-forecast targeting central banks typically rely on a model-based forecasting and policy analysis. They adopt a forecasting and policy analysis system (FPAS) that provides a more forward-looking approach that recognizes, and takes into account, the limitations of models. It allows a more rigorous decision-making process that avoids the short-term bias that can result from a less-structured, data-driven approach. This is achieved by designing and parameterizing a model to broadly reflect the MPC’s view of the economy and reliably track its performance. The main elements of FPAS are: (i) a model-based macroeconomic forecast with an endogenous future path for the short-term interest rate; (ii) incorporation of uncertainty into policy decision making; and (iii) an assessment of risks to monetary policy based on alternative scenarios, as well as other factors not captured by the model.
12. Monetary policy using FPAS can be illustrated with a model calibrated for Korea. This model is a new-Keynesian, Dynamic Stochastic General Equilibrium with model consistent expectations, which is used in some form by most IFT central banks. At its core are equations for the output gap, core inflation, the policy interest rate, and the exchange rate. The output gap is a function of its own lead and lag, foreign activity, and the deviation from their equilibrium level of the real longer-term interest rate and the real exchange rate. Core inflation is determined by an expectations-augmented Phillips curve in which the output gap drives short-term changes in the inflation rate. The exchange rate is linked to the interest rate through an uncovered interest rate parity condition. Additional equations define headline inflation; trade and financial linkages with the rest of the world; and the yield curve. Equations are linear except for the Phillips Curve, where the non-linearity captures the feature that the Philips Curve becomes quite flat when the negative output gap is large.
13. The model policy reaction function ensures that inflation eventually converges to the inflation target. It uses two alternative policy reaction functions. The first is a linear Taylor rule in which the policy interest rate responds to the deviation between the model forecast of inflation three-quarters ahead and the inflation target, and to the output gap. The second is a quadratic-loss policy reaction function that responds more strongly to larger shocks. It incorporates interest rate smoothing to make the adjustment in the policy rate more gradual.
14. The model serves two purposes. First, it provides a forecast from initial conditions, where there is an output gap and inflation differ from target, and illustrates how inflation converge back to target under alternative macro policies and monetary reaction functions. Second, it assesses macroeconomic risks by implementing scenarios in which shocks widen the output gap and cause deviation of inflation from target.
E. Illustrative Simulation of Alternative Monetary Policy Scenarios
15. The benefits of the FPAS approach can be illustrated by applying this model to the conditions that existed in 2013 when the recent easing cycle began. A comparison of the paths for the policy rate and the macro variables projected by the model to actual outcomes starting from 2013, indicates how monetary policy would have been different under FPAS. The Figure compares the historical evolution of the main variables and actual policy rate, including eight policy rate cuts totaling 200 basis points over three years, over this period, to the model’s projections. This is done for two policy reaction functions—the linear, inflation-forecast-based (IFB) reaction function and the non-linear loss-minimization approach (Figure 1).
Figure 1.Model Projections from 2013 for Two Reaction Functions Compared to History
Source: Authors’ calculations.
16. This comparison between actual and model outcomes can be made more realistic by incorporating shocks hitting the economy over this period. The model scenario starts in 2013 and assumes no new shocks; whereas, in reality, shocks hit the economy after 2013 on an ongoing basis and are reflected in the actual outcomes in the chart. To reflect this, the model can incorporate into the simulation the sequence of actual historical shocks that hit the Korean economy after 2013. With the post-2013 shocks added, the economy is much weaker than in the previous scenario and, in contrast to the previous scenario, convergence to the inflation target is not achieved until much later (and beyond the time scale shown in Figure 2). As a result, the model’s projected paths for the policy rate is lower where, after an initial series of rate cuts, the policy rate continues to be reduced gradually (in line with what actually occurred). Overall, the addition of these historical shocks to the simulation leads to a larger cumulative rate cut and exchange rate depreciation (that shown for the quadratic loss reaction function only). Of course, the final policy rate decision will take into account factors not reflected in the model.
Figure 2.Actual and Model Outcomes with Historical Shocks Added to the Scenario
Source: Authors’ calculations.
17. In a downside scenario, coordination of monetary and fiscal policy can produce better outcomes. A more expansionary fiscal policy allows a more moderate monetary easing and reduces the probability that monetary policy will be constrained by the ELB. This downside scenario shock is assumed to start in mid-2017 with a widening of the output gap by 1 percentage point, and a 0.5 percentage points decline in inflation (Figure 3). The shock was chosen to be large enough for the interest rates to hit the ELB when monetary policy alone is used. However, when combined with a fiscal expansion the central bank does not need to cut the policy rate to the ELB, and the period of low interest rates is shorter. This positive effect of fiscal policy is magnified by the fact that when the interest rate falls to the ELB, the fiscal multiplier is larger in the model as there is little or no crowding out effect from the fiscal expansion (which normally pushes up the interest rate).
Figure 3.Downside Scenario with Fiscal Backstop
Source: Authors’ calculations.
18. The reduction in the inflation target by a percentage point to 2 percent in January 2016 weakened the nominal anchor. Monetary policy can play a role rebuilding the credibility of the anchor more rapidly through the adoption of inflation-forecast targeting. This involves greater reliance on model-based forecasting and policy analysis to strengthen the management of expectations and as an input into overall policy decision. This can enhance the credibility of the official inflation target and, ultimately, monetary policy transmission. It involves providing greater forward guidance to shape inflationary expectations and foster private sector behavior that enhances the effectiveness of monetary policy.
19. This strengthening of the monetary policy framework involves enhancing communications. IFT involves providing a macroeconomic forecast of inflation and effective communicating the central bank’s reaction function – how it would react to shocks to achieve its inflation target. This also entails clarifying how policy will manage uncertainty related to the output gap, the equilibrium real interest rate, capital flow volatility and financial stability, and resolve any conflicts among objectives.
20. An effective, credible monetary policy cannot address all macroeconomic challenges facing Korea. Rather, it can foster robust growth with low inflation, providing a stable and predictable environment that allows other policies to work more effectively. These other policies play a complementary role (Gaspar et. al., 2016). Fiscal policy can reinforce the effectiveness of monetary policy, as illustrated by model scenarios. Structural policies can also support monetary policy by, for example, boosting potential growth. Macroprudential policies that ensure financial stability and limit the buildup of risks by preventing excessive credit growth to households, allow monetary policy to focus solely on the inflation target and remove incentive to deviate from this objective to address financial stability concerns.
Berg, A., P.Karam, and D.Laxton, 2006, “A Practical Model-Based Approach to Monetary Policy Analysis—Overview,”IMF Working Paper No. 06/80.
Ball, L., 2014, “The Case for a Long-Run Inflation Target of Four Percent,”IMF Working Paper No. 14/92.
Blanchard, O., 2014, “Where Danger Lurks,” Finance & Development, September 2014, Vol. 51, No. 3.
Bank of Korea, 2016, “Monetary Policy Direction for 2017,”December.
Clinton, K., C.Freedman, M.Juillard, O.Kamenik, D.Laxton, H.Wang, 2015, “Inflation-Forecast Targeting: Applying the Principle of Transparency,”IMF Working Paper No. 15/132.
Clinton, K., C.Freedman, M.Juillard, O.Kamenik, D.Laxton, H.Wang, T.Hlédik, T.Holub, D.Laxton, and H.Wang, 2017, “Czech Magic: Implementing Inflation-Forecast Targeting at the CNB,”IMF Working Paper No. 17/21.
Coenen, G., C.Erceg, C.Freedman, D.Furceri, M.Kumhof, R.Lalonde, J.Linde, A.Mourougane, D.Muir, S.Mursula, J.Roberts, W.Roeger, C.de Resende, S.Smudden, M.Trabandt, J.in‘t Veld, 2010, “Effects of Fiscal Stimulus in Structural Models,”IMF Working Paper 10/73.
Dooley, M., R.Dornbusch, and Y.C.Park, 2002, “A Framework for Exchange Rate Policy in Korea,”Korea Institute for International Economic Policy, Working Paper 02-02.
Freedman, C. and D.Laxton, 2009, “Inflation Targeting Pillars: Transparency and Accountability,”IMF Working Paper No. 09/262.
Freedman, C. and D.Laxton, M.Kumhof, D.Laxton, and J.Lee, 2009a, “The Case for Global Fiscal Stimulus,”IMF Staff Position Paper No. 09/04.
Freedman, C. and D.Laxton, M.Kumhof, D.Laxton, D.Muir and S.Mursula, 2009b, “Fiscal Stimulus to the Rescue? Short-Run Benefits and Potential Long-Term Costs of Fiscal Deficits,”IMF Working Paper No. 09/255.
Freedman, C. and D.Laxton, M.Kumhof, D.Laxton, D.Muir, and S.Mursula, 2010, “Global Effects of Fiscal Stimulus During the Crisis,”Journal of Monetary Economics, Vol. 57 (June), pp. 506–526.
Gaspar, V., M.Obstfeld, R.Sahay, D.Laxton, D.Botman, K.Clinton, R.Duval, K.Ishi, Z.Jakab, L. J.Mayor, C. L.Ngouana, T. M.Griffoli, J.Mongardini, S.Mursula, E.Nier, Y.Ustyugova, H.Wang, and O.Wuensch, 2016, “Macroeconomic Management When Policy Space Is Constrained: A Comprehensive, Consistent and Coordinated Approach to Economic Policy,”IMF Staff Discussion Note No. 16/09.
Laxton, D., D.Rose, and A.Scott, 2009, “Developing a Structured Forecasting and Policy Analysis System to Support Inflation-Forecast Targeting (IFT),”Bank of Canada WP/09/65.
Obstfeld, M., K.Clinton, O.Kamenik, D.Laxton, Y.Ustyugova, and H.Wang, 2016, “How to Improve Inflation Targeting in Canada”, IMF Working Paper No. 16/192.
Park, Y.C., and C.Wyplosz, 2009, “Inflation Targeting in Korea: Success, Good Luck or Bad Luck?” in Karasulu and Yang, 2009.
By R. Sean Craig (APD) and Hou Wang (RES), based on the forthcoming IMF Working Paper “Enhancing the Monetary Policy Framework in Korea” (2018), by Kevin Clinton, R. Sean Craig, Douglas Laxton, and Hou Wang.
Evidence is provided in the 2017 Korea Article IV Staff Report, Box 2, titled “Central Bank Experiences with Inflation-Forecast Targeting.”
Reported in “Estimation of Korean Economy’s Potential Growth Rate,” Bank of Korea Monthly Bulletin, August 2017. The estimates represent the views of the authors of the article and do not necessarily reflect the official view of the Bank of Korea.