Monetary policy under zero bound on short-term nominal interest rates : Critical appraisal on the ineffectiveness viewpoints(in Japanese)

1. Background There have been oppositions to and doubts on additional quantitative monetary easing after nominal short-term rates hit zero, mainly led by the central bankers and financial market participants. This paper critically examines the robustness of their views from various aspects. 2. Tools We present the empirical counter-arguments to the proponents of the above views. We utilize several tools including vector autoregressions (VARs), panel data analysis from major advanced countries, and long-run cointegration analysis, as well as historical and narrative approaches of the US Great Depression. We had to rely on this type of ad-hoc approaches since the opposition views use different approaches. This, however, enable us to view the current deadlock in Japan's monetary policy and its impact on economic activity and prices from historical and long-run perspectives. 3. Main results Monetary policy maintains various transmission channels even for the economy facing the zero bound on nominal rates. It affects real economic activity without heavily relying on the bank lending channel. The liquidity trap argument does not explain the resilient private consumption. While deflation and its expectation postpone consumption, nominal zero interest rates make the distinction ambiguous between the storage of value through financial assets and that through real assets, thereby stimulating consumption of certain durables and big-ticket items. We take this as an indication that the wealth effect from increases in liquid financial assets supports consumption. We believe that additional monetary easing will generate sufficient portfolio substitution among financial and real assets, when we reach an unobservable bliss point for the demand for money since we do not obtain utility from holding financial assets. It is contradiction to argue both that monetary policy is ineffective and that additional monetary easing will entail the risk of hyperinflation through damaging the credibility of the central bank. The direct underwriting of the government securities by the central bank, which is highly notorious, and the current scheme of the central bank's outright purchase of government securities do not differ in substantive ways. The argument linking the direct underwriting and hyperinflation does not always hold. 4. Concluding remarks We require a new type of monetary policy guideline (no matter how primitive and experimental) as we enter unexplored areas. We propose, as a rough guideline, an introduction of the "growth rate" targeting of monetary base, thereby influencing inflationary expectation, in order to recover positive nominal GDP growth. Under certain assumptions, the annual growth rate of 5 to 6% of M2+CD is required to achieve positive rates of inflation by 1 to 2% a year (measured by GDP deflator). It will require annualized monetary base growth of 12 to 22% with the outright purchases of securities by 1.3 to 2.1 trillion yen per month. To avoid a possible distortion the central bank's purchase of securities might cause under deflation, we propose the outright purchases to be proportional to the weights of securities held by the private financial and nonfinancial sectors. This comes roughly to the weights of 2 for government securities, 1 for foreign-currency denominated securities, and 1 for domestic equities (including Exchange-Traded Funds). The total annual purchases of these securities will be 15 to 25 trillion yen, still below the cyclically adjusted primary deficit of the general government (around 30 trillion yen), thus expected to limit the risk of rapid inflation.