This presentation will interrogate the claim made by financial journalists and observers that the Bank of England kept interest rates higher than necessary when it switched to inflation targeting in the 1990s. The framework is broad enough to evaluate any central bank pursuing inflation targeting.
Dissertation thesis presentation by Pranjal Rawat | General Economics | 2016 - 8 | Access link to the presentation files
Inflation Forecast Targeting: Lessons from UK
“…the real influence of monetary policy is less the effect of any individual monthly decision on
interest rates and more the ability of the framework of policy to condition inflation expectations.”1
~ Mervyn King
Inflation Forecast Targeting (IFT) is a comprehensive framework for monetary policy that commands worldwide respect. This article reviews the lessons that can be drawn from the UK experience with IFT. The UK experience is important because it was very successful in taming inflation. It’s success, however, led to other problems. Indian monetary policy, having adopted IFT in 2016, would do well to pay attention to such valuable institutional experiences.
Monetary & Exchange Rate Targeting
From 1970 to 1990, the UK struggled to deliver low and stable inflation without compromising on growth and unemployment.
From the mid-1970s to mid-1980s, the Bank of England (BoE) conducted monetary targeting. Interest rates were raised to target a slower money growth. In practice they used many target aggregates (from M3 to M0) and allowed substantial overshooting without reason. There was also irregular scheduling of announcements. The problem worsened when the money growth-inflation relationship broke down.
The result was that interest rates had to be raised considerably to bring inflation down marginally. And there was a huge cost to this. The UK suffered a severe recession in 1980. Output was highly volatile, and unemployment rose sharply.2
So in 1987, the BoE switched to Exchange Rate targeting. British monetary policy mirrored German monetary policy, and keeping the Pound fixed to the Deutsch mark was the commitment device. This was first done secretly, and then publicly only after 1990. So in 1988, when the Bundesbank raised rates, so did BoE.
The BoE struggled and hiked rates a lot, to keep the Pound within the target range. On 16th September 1992, currency speculators took advantage of this. They began to borrow Pounds, only to sell them off in large volumes. The BoE was unable to prevent sudden depreciation, even after selling immense forex and announcing more rate hikes.3 Again the cost of reducing inflation was a recession, in 1990, and currency crisis.
Inflation Forecast Targeting
Stung by the crisis, the UK initiated institutional reforms4. First, it explicitly defined the sole responsibility of monetary policy to be ‘price stability’. The annual growth in Retail Price Index Excluding Mortgage Payments (RPIX) would be maintained within a band of 1 % to 4 %. And attain 2.5% inflation p.a. or lower in 5 years.
Second, the BoE abandoned its history of secrecy. A quarterly Inflation Report was published which would articulate the BoE’s assessments, the rationale behind decisions and future road-map. The Report would also publish forecasts of inflation, which would keep the bank accountable. Third, regular monthly meetings between the Chancellor, BoE and Treasury was set up and its minutes published. Fourth, policy rates would be set to keep the forecast of inflation within a target range.
The results were fairly immediate. RPIX inflation fell to 2.5% by 1994. And the policy rate was actually brought down, from a high of 15% in 1990 to a low of 5% by 1994. Thus, GDP growth rates rebounded. However, unemployment rates were still very high.
In 1997, the Monetary Policy Committee was formed. It replaced the Chancellor’s role in setting rates. The Governor had to write an open letter for deviations larger than 1 percentage point.
After 1997, the results were quite good5. RPIX inflation stabilised near 2.5%, even after the BoE switched to targeting CPI inflation at 2% in 2003. Unemployment rates fell from a high of 10% in 1993:Q2 to below 4% by 1999:Q3. Sometimes they were as low as 2.4%. There were no recessions. Rates held stable and low at around 5%.
Macro Indicators Before and After FIT
Expectations & Credibility
All economic decisions like investing, pricing and wage negotiation depend on inflation expectations. And high inflation expectations are self-fulfilling prophesies. Workers expecting higher inflation will refuse to work at lower wages, putting pressure on employers to raise prices. Firms anticipating higher inflation will raise product prices to keep up. Similarly, low inflation expectations lead to low inflation. Policy must, therefore, bring down inflation expectations.
Inflation expectations depend on expectations about future monetary policy conduct. If future monetary policy is not credible in its pursuit of price stability, then no rational agent would expect inflation to be low.
The problem with monetary and exchange rate targeting was that their pursuit of price stability was not very credible. There was secrecy, an explicit focus on growth & unemployment, the shifting of goal-posts, the lack of accountability, dovish central banking with weak resolve. There was also a belief in a long-term trade-off between inflation and growth. All of these lead to inflation bias, where average inflation expectations and inflation are higher than what is socially desired.
Data tells us that inflation expectations were very high before IFT6. Inflation fell after 1992 due to the explicit inflation target and transparency & accountability measures. But unemployment took longer to fall. In 1997, when BoE attained operational independence, unemployment fell sharply and inflation expectations were anchored to a target.
FIT worked because it convinced people that BoE would be fully committed to battle inflation and was incentivised to take this battle to the very end. The rollout of IFT was also very gradual, unlike the earlier application of monetarist policies.
The forecast of inflation is generated by the best forecasting models. By construction, their relationship with inflation cannot break down. The same cannot be said for monetary aggregates and exchange rates. In fact, it can be shown that even when monetary aggregates have stable relation to inflation; this information can be easily incorporated into the forecast of inflation, by incorporating monetary aggregates into the forecasting model.
So inflation and inflation expectations fell, without the BoE actually having to increase interest rates very much. Institutional reform was enough to make the private sector believe that inflation would be controlled.
What about other factors like oil shocks? It’s clear that massive oil shocks did hit the UK economy in the 2000s. But this did not lead to a rise in average inflation because inflation expectations were anchored.
Speculation and Bubbles
For a long time, IFT worked well but it was a victim of its own success7. It’s ability to obtain low interest rates, even while maintaining low inflation, led to complacent lending. Consumers and investors made use of cheap loans to invest recklessly in assets. This led to asset price bubbles, most notably in the real estate markets. Financial technologies also evolved, and a wide variety of financial products flooded the market. A lack of intelligent regulation failed to keep bubbles from bursting. All this and more led to the financial crisis and the Great Recession.
And because interest rates were already low, they couldn’t be dropped lower to revive the economy. The problem of the zero lower bound arose. Thus, rates were brought down to zero and conventional IFT was abandoned for quantitative easing. Even then, the low inflation environment was not very encouraging to business and the recovery from recession was prolonged.
Clearly, IFT has its own problems. Low interest rate environments can lead to unintended problems. For instance, even while NPA ratios of Indian banks cannot guide monetary policy under inflation targeting, there are important spillover effects of monetary policy on the health of the banking sector. For instance, there is some evidence for a pro-cyclicality relationship between credit growth and NPA accumulation (Mishra & Dhal 2012, Chavan & Gombacorta 2016). If this is correct, then when IFT in India achieves a low interest rate environment it is likely that rapid credit expansion could push NPA ratios to unsustainable levels. That would raise the probability of a crisis in the banking industry.
The wrong way to resolve this is to confuse instruments with targets for which they are not intended. Which is to try and use policy repo rates to target both inflation and maintain NPA ratios at sustainable levels. Such an approach will destabilize the inflation-targeting framework because one instrument cannot attain two separate objectives.
The right way to resolve this is to institutionalize wide-ranging and bold reforms in the banking sector with a particular focus on macro-prudential norms. A close examination of the experience of China and Brazil who successfully brought down NPAs through this method is warranted.
References
1. King, Mervyn. "Monetary Policy: Practice Ahead of Theory (Mais Lecture 2005)." Lecture Delivered to the Cass Business School, City University, London 17 (2005).
2. King, Mervyn. "Monetary Policy in the UK." Fiscal Studies 15, no. 3 (1994): 109-128.
3. Inman, Phillip. "Black Wednesday 20 years on: How the Day Unfolded." The Guardian. September 13, 2012. Accessed December 13, 2017. https://www.theguardian.com/business/2012/sep/13/black-wednesday-20-years-pound-erm.
4. King, Mervyn. "Changes in UK monetary policy: Rules and discretion in practice." Journal of Monetary Economics 39, no. 1 (1997): 81-97.
5. King, Mervyn A. "What has Inflation Targeting Achieved?." In The Inflation-Targeting Debate, pp. 11-16. University of Chicago Press, 2004.
6. King, Mervyn. "The Institutions of Monetary Policy." the American Economic review 94, no. 2 (2004): 1-13.
7. King, Mervyn. "Twenty Years of Inflation Targeting." Stamp Memorial Lecture, London School of Economics, London, October 9 (2012).
Appendix A: Graphs
コメント