Unconventional Policy Tools

Once regarded as an historical relic of the Great Depression, the significance of the zero lower bound (ZLB) on interest rates as a constraint on the effectiveness of monetary policy has received much attention as a result of Japan's experience with deflation between 1999 and 2006 and concerns about the risk of deflation in the United States (US) following the collapse of the information technology bubble in 2000. Ironically, the seeds of the current recession in the US could partly be traced to the concerns of the US Federal Reserve about avoiding deflation in the early part of this decade, which led it to adopt an overly easy monetary stance. Although this bias toward easing was aimed at stabilizing the consumer price index (CPI), it ended up destabilizing housing and other asset prices, which contributed to the development of the housing bubble there. Now the US and much of Western Europe have fallen into liquidity traps as well. This can be seen most obviously by the breakdown of the traditional relationship between the monetary base (M0) and broad money (M2) (see Figure 2 [ PDF 15.2KB | 1 page ]). In other words, the money multiplier has broken down.

Much of the current literature on unconventional monetary policy can be traced back to Krugman, Dominquez, and Rogoff (1998), which focused on the problem that deflation prevents the real interest rate from falling enough to achieve full employment. Krugman and others argued that, in principle, the central bank must offset this by trying to raise the market's expectations about future inflation in order to bring down the real interest rate sufficiently to stimulate aggregate demand.

Bernanke and Reinhart (2004) divided unconventional monetary policy tools into three main categories:

1. providing assurance to investors that short rates will be kept lower in the future than they currently expect (“commitment effect”);
2. increasing the size of the central bank's balance sheet beyond the level needed to set the short-term policy rate at zero ("quantitative easing"); and
3. changing the relative supplies of securities in the marketplace by altering the composition of the central bank's balance sheet (“qualitative easing” or “credit easing”).

The central bank has another powerful option, namely, buying foreign currency assets in order to depress the value of the country's foreign exchange rate and thereby stimulate export demand. This effect was analyzed by Svensson (2001), among others, and could be particularly powerful for a small open economy. However, the stigma associated with adopting “beggar-thy-neighbor” policies appears to have effectively discouraged central banks from adopting such policies during economic downturns. There are plenty of recent examples of central banks intervening to maintain a stable exchange rate or to slow currency appreciation during an expansion phase, but no obvious examples of intervention to engineer currency depreciation as a macroeconomic stabilization tool during an economic downturn. In the remainder of this section, I examine the three categories of unconventional measures described above.

Commitment effect: A large literature has developed around the first category, which generally is referred to as the “commitment” or “policy duration” effect. The basic idea is simple—even though the central bank may set the very short-term rate, normally the overnight interbank rate, at zero, the market still has considerable uncertainty about the future development of monetary policy. This is reflected in the yield curve, since longer-term rates essentially reflect the market's expected future path of short-term rates plus a risk premium. Therefore, if the central bank can persuade the market that it will keep the policy rate lower than the market would expect otherwise, this should cause longer-term rates to fall, thereby stimulating the economy. This type of policy has been analyzed theoretically by a number of authors, including Svensson (2001) and Eggertsson and Woodford (2003).

Typically, the central bank commits to maintain its policy interest rate at zero for a certain period. This commitment could be conditional or unconditional, but normally is conditional because a central bank cannot reasonably be expected to ignore future developments. In particular, it would be normal for the central bank to start raising interest rates once the economy has recovered and inflation has begun to pick up, so the central bank might commit to keep rates at zero until these conditions were achieved. The first instance of such a commitment in recent times was the declaration by the Bank of Japan (BoJ) in April 1999 that it would maintain its zero-interest-rate policy until “deflationary concerns were dispelled” (Okina and Shiratsuka 2004: 75–76). In May 2001, the Bank of Japan took a more refined approach by promising that it would keep its policy rate at zero until consumer price inflation “stably” registered zero percent or positive year-on-year growth (Bank of Japan 2001). It further clarified its definition of what the end of deflation meant in October 2003 (Bank of Japan 2003). In the US, the August 2003 statement of the Federal Open Market Committee that "policy accommodation can be maintained for a considerable period" is another example of a commitment by policymakers (US Federal Reserve Board 2003). Indeed, the Fed is now using similar language, as its policy statement in December 2008 noted that “economic conditions are likely to warrant exceptionally low levels of the federal funds rate for an extended period” (US Federal Reserve Board 2008).

Many, including Reifschneider and Williams (2000), have proposed that the commitment to keep rates at zero should be maintained well beyond the time when inflation turns positive. This would effectively lower the market's expectation of real interest rates, thereby imparting a greater stimulus to the economy and minimizing the losses of output resulting from the economy being stuck at the zero bound during periods of inflation. Many variations of “backward-looking” policy rules for the inflation target, price-level target, or modified Taylor Rules have been proposed to minimize output losses. They generally have the feature that, the greater the cumulative loss in output due to deflation, the more the policy target must adjust, in terms of higher target inflation rate or price level, in order to compensate for this. However, there are no recent examples of central banks implementing such rules, so they will not be examined in this study.

Quantitative easing: Another form of easing is to expand the size of the central bank's balance sheet by increasing the size of reserve deposits—current account balances (CABs)—beyond the level that is required to bring the overnight funds rate to zero. (The monetary base consists of both cash in circulation and reserve deposits, but the central bank can only directly affect the level of reserve deposits.) This is referred to as “quantitative easing” (QE) and, according to Bernanke and Reinhart's typology, focuses on the liabilities side of the central bank's balance sheet. Possible channels of impact of such a policy include (i) the portfolio balance effect, i.e., if money is an imperfect substitute for other financial assets, the rise in money holdings leads investors to shift toward other assets, thereby raising their value and stimulating final demand; (ii) providing a clearer signal of the central bank's commitment to keep the policy rate low; and (iii) a permanent increase in the money supply could reduce the expected value of government debt servicing costs, thereby reducing the expected value of future tax payments. The first effect was investigated by Goodfriend (2000) in detail. The third effect was investigated by scholars including Auerbach and Obstfeld (2005).

The magnitude of the portfolio balance effect can be influenced significantly by the interest rate that the central bank pays to banks on their reserve deposits at the central bank. If the central bank pays a positive interest rate on reserve deposits, this will discourage banks from shifting out of excess reserves into other assets such as loans. Paying interest on reserve deposits is very close to sales of bills by the central bank as a funds-absorbing operation to tighten money market conditions. On the other hand, if the central bank pays a zero or even a negative interest rate, this would encourage re-intermediation. Goodfriend (2000) analyzed how the central bank could push nominal interest rates below zero throughout the economy by paying negative interest rates on reserve deposits. On 1 July 2009, the Swedish Riksbank actually cut the interest rate it pays on reserve deposits to minus 0.25%, the only recent instance of a central bank doing so. US Federal Reserve Chairman Ben S. Bernanke observed that recent US legislation to allow the Fed to pay interest on reserve deposits gives the Fed greater flexibility to reduce its balance sheet when it needs to implement its “exit strategy” for tightening monetary policy (Bernanke 2009b). The last point is discussed further below.

The main theoretical objection to quantitative easing is that, at zero interest rates, money and short-term paper are perfect substitutes, so changes in the level of current account balances simply represents shifts in holdings of assets that are essentially the same, and hence should have no real economic impact. However, there is some evidence in favor of the portfolio balance effect, which is discussed below.

The main example of quantitative easing was its implementation by the Bank of Japan in April 2001, when it shifted its policy target from the overnight call rate to the level of CABs. This policy was maintained until March 2006. Most recently, the Bank of England adopted quantitative easing on 5 March 2009, although the bank rate is still positive.

One drawback of quantitative easing and, indeed, of the zero-interest-rate policy (ZIRP) is the distortion of the functioning of the money market due to the very low level of interest rates resulting from the fact that money market brokers cannot cover their costs. Indeed, balances in Japan's call market dropped dramatically by almost half during the operation of the QE policy, while balances in the Euroyen market fell almost 90%. As a result, during the current global financial crisis, the BoJ has kept the overnight call rate at 0.1%, just high enough to cover the costs of the money market brokers. Other central banks have tended to keep the policy rate sufficiently above zero to preserve the functioning of money markets as well.

Qualitative easing: The third set of policies is aimed at varying the mix of assets held by the central bank, and is referred to as “qualitative easing” or “credit easing.” The basic idea is that operations to change the shares of various kinds of assets held by the private sector will lead to changes in their relative prices, and thereby have implications for real economic activity. For example, if the central bank increases its outright (permanent) purchases of long-term government bonds, this could be expected to reduce long-term bond yields and stimulate the economy. Qualitative easing also includes direct lending to market participants in cases where the normal transmission mechanism breaks down. In this case, policies are focused on reducing credit market spreads and improving the functioning of private credit markets more generally. Like quantitative easing, qualitative easing generally involves an increase in the size of the central bank's balance sheet, but the focus is on the mix of assets, not the level of bank reserves (liabilities). This is particularly relevant during the current global financial crisis, where credit spreads have been much wider and credit markets more dysfunctional in the US and other countries than was the case in Japan during Japan's experiment with quantitative easing.

In a recent speech (Bernanke 2009a), US Federal Reserve Chairman Bernanke distinguished between three kinds of qualitative easing activity:

1. the central bank's traditional role of provision of short-term liquidity to sound financial institutions, i.e., “lender of last resort” activity;
2. provision of liquidity directly to borrowers and investors in key credit markets; and
3. purchase of longer-term securities for the Fed's portfolio.

All of these measures could be seen as responses to malfunctioning of credit markets due to severe market concerns about capital adequacy and bankruptcy risk.

The first category starts with traditional borrowing at the discount window, which is not “unconventional,” although it has long played a secondary role to open-market operations. The most straightforward kinds of easing involve relaxing the criteria for the kinds of borrowers or types of collateral that qualify for open-market operations, or extending the period of such operations. Financial market stresses led to the creation of a number of new programs in this category. In the US, this includes a number of new credit facilities for auctioning credit that were responses to stresses in the interbank funding market: the Term Auction Facility (TAF); Term Securities Lending Facility (TSLF); Primary Dealer Credit Facility (PDCF); and bilateral currency swap agreements with 14 foreign central banks, including the Bank of England, the European Central Bank, the Bank of Japan, the Bank of Korea, Monetary Authority of Singapore, and the Swiss National Bank (see Appendix 1 [ PDF 8.5KB | 1 page ] for a brief description of some of these programs and Appendix 2 [ PDF 10.1KB | 2 page ] for a chronology of announcements by the Fed and other central banks). The TAF was aimed at solving the stigma problem related to banks borrowing from the Fed, while the TSLF and PDCF provided comparable facilities for primary dealers. The swap agreements were aimed at easing shortages of US dollars in overseas markets. These loans are viewed as having very low risk, since they generally are over-collateralized and with recourse. Also, the foreign exchange swap agreements are made with other central banks, where there is a high degree of mutual trust.

The second category of policies was aimed at other markets besides the interbank market, including the commercial paper market, the asset-backed securities market, and money market funds, which also showed increased signs of stress. In the US, these new programs included Asset-Backed Commercial Paper Money Market Fund Liquidity Facility (AMLF); Commercial Paper Funding Facility (CPFF); Money Market Investor Funding Facility (MMIFF); and Term Asset-Backed Securities Loan Facility (TALF). The measures for assetbacked securities were aimed at AAA-rated securities collateralized by student loans, auto loans, credit card loans, and loans guaranteed by the Small Business Administration (jointly with the US Treasury to cover risk). Again, credit risk is seen as very low in both programs. In particular, the TALF program requires that loans be over-collateralized and is further protected by capital provided by the Treasury.

Under the third category, the Federal Reserve announced on 18 March 2009 that it will purchase cumulative amounts of up to US$1.25 trillion of agency mortgage-backed securities and up to US$200 billion of agency debt by the end of the year, and up to US$300 billion of longer-term Treasury securities over the following six months (US Federal Reserve Board 2009a). The principal goal of these programs is to lower the cost and improve the availability of credit for households and businesses. On 19 January 2009, the United Kingdom (UK) Treasury specified the following types of sterling-denominated assets as eligible for purchase under the Bank of England's Asset Purchase Facility: commercial paper, corporate bonds, bonds issued under the UK's credit guarantee scheme, syndicated loans, and asset-backed securities “created in viable securitization structures” (Her Majesty's Treasury 2009).

Figure 3 [ PDF 25KB | 1 page ] shows the trend of these different kinds of assets in the Fed's balance sheet. There has been a clear shift toward the third category since the Fed's statement on 18 March 2009, as the share of outright purchases in total assets has risen from 37% to 57% as of 10 June. Notably, the overall level of Fed assets has been relatively constant, as the rise in long-term holdings has been offset by substantial declines in short-term liquidity-related assets such as central bank liquidity swaps and the CPFF as funding pressures have eased in these segments. These trends suggest that the Fed already has its exit strategy well in mind, as will be discussed below.

During the period of the QE policy, the Bank of Japan progressively increased the level of outright Japanese government bond (JGB) purchases from ¥400 billion per month to ¥600 billion on 14 August 2001; to ¥800 billion on 19 December 2001; to ¥1,000 billion on 28 February 2002; and finally to ¥1,200 billion on 30 October 2002. This level was maintained until 19 December 2008, when the Bank of Japan announced that the amount of outright purchases of JGBs would be increased to ¥1.4 trillion yen, and then finally to ¥1.8 trillion yen on 18 March 2009. It also expanded the range of JGBs accepted in outright purchases (30- year bonds, floating-rate bonds, and inflation-indexed bonds were added to the list of eligible JGBs) In addition, in order to prevent the remaining maturities of JGBs purchased from becoming too short or too long, it introduced a scheme to purchase JGBs from specific maturity segments (maturity segments are defined as 1 year or less, more than 1 year through 10 years, and more than 10 years). The BoJ also announced on 19 February 2009 that it would commence outright purchases of corporate bonds.

Table 1 [ PDF 10.2KB | 1 page ] shows a comprehensive list of qualitative easing measures adopted by various central banks during the current crisis. The breakdown follows that of Bernanke (2009a) mentioned earlier. Measures aimed at easing conditions in interbank markets were more numerous than those aimed at influencing credit markets, while those aimed at influencing broader financial conditions were rarest. This presumably reflects the relative unconventionality of the three stages, as central banks, being conservative, tended to favor modest steps in the direction of unconventionality. All banks conducted exceptional longterm operations, broadened eligible collateral, and participated in foreign exchange (FX) swap lines with the Fed. For example, on 2 December 2008, the Bank of Japan announced that it would ease the criterion on credit ratings from “A-rated or higher” to “BBB-rated or higher.” Also, the Bank of Japan introduced, as a measure to enhance flexibility in fundssupplying operations collateralized by corporate debt, a new operation that provides funds over the fiscal year-end at an interest rate equivalent to the target for the uncollateralized overnight call rate. It also included the Development Bank of Japan Inc. as a counterparty in operations such as commercial paper (CP) repo operations. On 22 January 2009, it announced acceptance of debt instruments issued by real estate investment corporations as eligible collateral as well. On 19 February, government-guaranteed dematerialized commercial paper was included in eligible collateral, and the range of Japanese government securities offered in the security lending facility was broadened. On 7 April 2009, it also accepted loans on deeds to municipal governments as eligible collateral.

Regarding steps to influence credit market conditions, the most commonly taken steps were aimed at CP funding, followed by asset-backed securities and corporate bonds. Only the Bank of Japan and the Swiss National Bank purchased other non-public-sector securities such as equities. Among steps to influence broader market conditions, outright purchases of public sector securities were more common than those of private sector securities, but only three central banks did even this.

Download this Paper [ PDF 294.1KB| 34 pages ].

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.