QE. As noted, because of the importance of the signaling channel, it is essential that asset purchases and interest rate policy be closely integrated, and that in particular the central bank be clear about its planned sequencing of the introduction and withdrawal of its various tools The portfolio balance channel depends on the premise that securities are imperfect substitutes in investors' portfolios, reflecting differences in liquidity, transactions costs information, regulatory restrictions, and the like Imperfect substitutability implies that changes in the net supply of a security affect asset prices and yields, as investors must be induced to rebalance their portfolios( Bonis et al., 2017). In principle, the two channels of QE can be distinguished by the fact that the signaling channel operates by affecting expectations of future policy rates while the portfolio balance channel works by changing term and risk premiums There have been many studies of the effectiveness of QE, mostly through event studies of the impact of QE announcements on interest rates and asset prices; for surveys, see, e.g, Gagno (2016b), Bhattarai and Neely(2016), and Williams(2014 ). There is not a large sample of QI programs to study, and econometric identification of the unexpected(and thus not fully discounted) components of QE announcements is difficult. Consequently, disagreements remain among researchers about the magnitude and persistence of QE effects and about the relativ mportance of the two primary channels of effect. Nevertheless, the strong view that QE is ineffective has been pretty decisively rejected. There appears instead to be a broad consensus that QE has proven a useful tool, with demonstrable effects on financial conditions. QE has been found to have significant effects on both rate expectations and term premiums, suggesting that both the signaling and portfolio balance channels are operative(Bauer& rudebusch, 2013 Huther et al., 2017). And, although showing direct links to macroeconomic outcomes is not straightforward, the experiences of the U.S., U. K. Japan, and Europe all suggest that the use of large-scale QE has been followed, over the subsequent couple of years, by strengthenin aggregate de emand and Improv ed economic performance(Engen et al., 2015) S For example, Gagnon(2016b), Table 1, reports 18 estimates from 16 studies of the effects of QE bond purchases on bond yields. For U.S. data, the median effect of a hypothetical program sized at 10 percent of GDP on 10-year yields is 82 basis points. Using the conventional rule of thumb that a 10 basis point reduction in the 10-year yield is about equivalent to a 25 basis point cut in the federal funds, thats roughly equal to 200 basis points of funds rate reductions for a program of that size. (The Feds program was considerably larger than 10 percent of GDP. According to Gagnon's survey, the median effect of purchases programs on the term premium component of 10-year yields only is 44 basis points, suggesting that both signaling effects and portfolio balance effects operate 6 It is true that QE has not been sufficient in a number of those cases to return inflation to target. However, the weak link in the causal chain appears to be in the influence of declining slack on inflation, not the effect of monetary
10 QE. As noted, because of the importance of the signaling channel, it is essential that asset purchases and interest rate policy be closely integrated, and that in particular the central bank be clear about its planned sequencing of the introduction and withdrawal of its various tools. The portfolio balance channel depends on the premise that securities are imperfect substitutes in investors’ portfolios, reflecting differences in liquidity, transactions costs, information, regulatory restrictions, and the like. Imperfect substitutability implies that changes in the net supply of a security affect asset prices and yields, as investors must be induced to rebalance their portfolios (Bonis et al., 2017). In principle, the two channels of QE can be distinguished by the fact that the signaling channel operates by affecting expectations of future policy rates while the portfolio balance channel works by changing term and risk premiums. There have been many studies of the effectiveness of QE, mostly through event studies of the impact of QE announcements on interest rates and asset prices; for surveys, see, e.g., Gagnon (2016b), Bhattarai and Neely (2016), and Williams (2014). There is not a large sample of QE programs to study, and econometric identification of the unexpected (and thus not fully discounted) components of QE announcements is difficult. Consequently, disagreements remain among researchers about the magnitude and persistence of QE effects and about the relative importance of the two primary channels of effect. Nevertheless, the strong view that QE is ineffective has been pretty decisively rejected. There appears instead to be a broad consensus that QE has proven a useful tool, with demonstrable effects on financial conditions.5 QE has been found to have significant effects on both rate expectations and term premiums, suggesting that both the signaling and portfolio balance channels are operative (Bauer & Rudebusch, 2013: Huther et al., 2017). And, although showing direct links to macroeconomic outcomes is not straightforward, the experiences of the U.S., U.K. Japan, and Europe all suggest that the use of large-scale QE has been followed, over the subsequent couple of years, by strengthening aggregate demand and improved economic performance (Engen et al., 2015). 6 5 For example, Gagnon (2016b), Table 1, reports 18 estimates from 16 studies of the effects of QE bond purchases on bond yields. For U.S. data, the median effect of a hypothetical program sized at 10 percent of GDP on 10-year yields is 82 basis points. Using the conventional rule of thumb that a 10 basis point reduction in the 10-year yield is about equivalent to a 25 basis point cut in the federal funds, that’s roughly equal to 200 basis points of funds rate reductions for a program of that size. (The Fed’s program was considerably larger than 10 percent of GDP.) According to Gagnon’s survey, the median effect of purchases programs on the term premium component of 10-year yields only is 44 basis points, suggesting that both signaling effects and portfolio balance effects operate. 6 It is true that QE has not been sufficient in a number of those cases to return inflation to target. However, the weak link in the causal chain appears to be in the influence of declining slack on inflation, not the effect of monetary
Controversies about Qe have focused less on whether the medicine works and more on the possible side effects. Many dark warnings accompanied the introduction of QE programs by major central banks after the financial crisis. In a memorable example, the Republican leadership of the U. S Congress wrote to the Fed in November 2010 to express concerns about further asset purchases. Their letter argued that"such a measure introduces significant uncertainty regarding the future strength of the dollar and could result both in hard-to-control long-term inflation and potentially generate artificial [sic] asset bubbles that could cause further economic disruptions(Herszenhorn, 2010). The Congressional leaders also were worried about foreign criticism of the Feds actions, noting that"any action.. that impairs U. S. trade relations at a time when we should be fighting global trade protection measures will only further harm the global economy and could delay recovery in the United States. " As the legislators noted, there was indeed foreign criticism of Fed plans for more QE, including from Brazilian finance minister Guido mantega, who argued that the Feds actions presaged a"currency war, and german finance minister Wolfgang Schauble, who reportedly called the policy""( Garnham Wheatley, 2010; Atkins, 2010). A subsequent letter from conservative economists and market participants echoed the themes of the Congressional letter, warning against"currency debasement and inflation "and adding that QE could"distort financial markets and greatly complicate future Fed efforts to normalize monetary policy"(Wall Street Journal, 2010). Less well remembered is that, in September 201 1 the Republican Congressional leadership wrote a similar, follow-up letter, adding the concern that QE could"promote borrowing by overleveraged consumers"(Wall Street Journal, 2011). Of course, these themes were staples of Vall Street Journal and Financial Times op-eds throughout the period I think it's fair to say that these warnings, and many more like them, have not proved prescient. Certainly there has been no massive upsurge in inflation--quite the opposite, of course-or a collapse in the dollar, as predicted by proponents of crude monetarism(of a type certainly, that Milton Friedman would never have endorsed ). Without a sustained decline in the dollar, and with a stronger U. S economy providing increased demand for imports, Mantegas concern about a currency war also proved baseless. Household leverage has not risen as the policy(including QE)on aggregate demand. The apparent flatness(or downward shift)of the Phillips curve is a problem for any macroeconomic policy aimed at raising inflation. See for example Taylor and Ryan(2010
11 Controversies about QE have focused less on whether the medicine works and more on the possible side effects. Many dark warnings accompanied the introduction of QE programs by major central banks after the financial crisis. In a memorable example, the Republican leadership of the U.S. Congress wrote to the Fed in November 2010 to express concerns about further asset purchases. Their letter argued that “such a measure introduces significant uncertainty regarding the future strength of the dollar and could result both in hard-to-control long-term inflation and potentially generate artificial [sic] asset bubbles that could cause further economic disruptions” (Herszenhorn, 2010). The Congressional leaders also were worried about foreign criticism of the Fed’s actions, noting that “any action….that impairs U.S. trade relations at a time when we should be fighting global trade protection measures will only further harm the global economy and could delay recovery in the United States.” As the legislators noted, there was indeed foreign criticism of Fed plans for more QE, including from Brazilian finance minister Guido Mantega, who argued that the Fed’s actions presaged a “currency war”, and German finance minister Wolfgang Schauble, who reportedly called the policy “clueless” (Garnham & Wheatley, 2010; Atkins, 2010). A subsequent letter from conservative economists and market participants echoed the themes of the Congressional letter, warning against “currency debasement and inflation” and adding that QE could “distort financial markets and greatly complicate future Fed efforts to normalize monetary policy” (Wall Street Journal, 2010). Less well remembered is that, in September 2011 the Republican Congressional leadership wrote a similar, follow-up letter, adding the concern that QE could “promote borrowing by overleveraged consumers” (Wall Street Journal, 2011). Of course, these themes were staples of Wall Street Journal and Financial Times op-eds throughout the period.7 I think it’s fair to say that these warnings, and many more like them, have not proved prescient. Certainly there has been no massive upsurge in inflation—quite the opposite, of course—or a collapse in the dollar, as predicted by proponents of crude monetarism (of a type, certainly, that Milton Friedman would never have endorsed). Without a sustained decline in the dollar, and with a stronger U.S. economy providing increased demand for imports, Mantega’s concern about a currency war also proved baseless. Household leverage has not risen as the policy (including QE) on aggregate demand. The apparent flatness (or downward shift) of the Phillips curve is a problem for any macroeconomic policy aimed at raising inflation. 7 See for example Taylor and Ryan (2010)
ond Congressional letter predicted; indeed, household debt and interest burdens have fallen significantly since the crisis Concerns about asset bubbles have been especially persistent(although these would appear to relate more to accommodative monetary policies in general than to Qe in particular) To be clear, there is no doubt that monetary policy affects the prices of stocks and other assets indeed, those effects are an important vehicle of monetary transmission. The intended effects of monetary easing on asset prices work through fundamentals, including the reduced discounting of future returns implied by lower interest rates, expectations of stronger economic performance, and moderate increases in risk-bearing capacity. Asset price increases due to those fundamental causes are desirable and pose no significant risks to economic or financial stability. Concern about bubbles is therefore properly focused on asset price increases that significantly exceed what can be justified by fundamentals. Claims that Qe has generated asset bubbles in this relevant sense are difficult or impossible to disprove; and, of course, at some point there will inevitably be a downward correction in asset prices, as has happened periodically in the past However, it's been seven years since the first Congressional letter, and the Fed stopped purchasing securities three years ago(although other central banks have continued it), so if QE has generated bubble dynamics we can at least conclude that some pretty long lags are involved What about other critiques? The claim that Qe distorts"financial markets, raised in the letter from economists and market participants, is heard fairly often. It's not clear exactly what that means. The goal of QE, and of monetary policy generally, is to set financial conditions consistent with full employment and stable prices, which can be thought of as trying to undo the economic distortions arising from price and wage stickiness, monopolistic competition, credit market frictions, and the like. In this respect appropriate monetary policy is"un-distorting: in particular, allocations under active monetary policy should be closer rather than further from the competitive, free-market, flexible-price ideal a possible rationalization of the"distortion"claim is that QE works, at least in part, by affecting term premiums(and through them, the whole gamut of asset prices ). From a market participants point of view, when QE is active it feels as if government(central bank) decisions rather than private-sector fundamentals, are setting asset prices. Moreover, in such situations it may appear that the highest returns go to the best Fed-watchers, rather than to those whose expertise is in evaluating economic fundamentals. Some frustration with this state of affairs on
12 second Congressional letter predicted; indeed, household debt and interest burdens have fallen significantly since the crisis. Concerns about asset bubbles have been especially persistent (although these would appear to relate more to accommodative monetary policies in general than to QE in particular). To be clear, there is no doubt that monetary policy affects the prices of stocks and other assets; indeed, those effects are an important vehicle of monetary transmission. The intended effects of monetary easing on asset prices work through fundamentals, including the reduced discounting of future returns implied by lower interest rates, expectations of stronger economic performance, and moderate increases in risk-bearing capacity. Asset price increases due to those fundamental causes are desirable and pose no significant risks to economic or financial stability. Concern about bubbles is therefore properly focused on asset price increases that significantly exceed what can be justified by fundamentals. Claims that QE has generated asset bubbles in this relevant sense are difficult or impossible to disprove; and, of course, at some point there will inevitably be a downward correction in asset prices, as has happened periodically in the past. However, it’s been seven years since the first Congressional letter, and the Fed stopped purchasing securities three years ago (although other central banks have continued it), so if QE has generated bubble dynamics we can at least conclude that some pretty long lags are involved. What about other critiques? The claim that QE “distorts” financial markets, raised in the letter from economists and market participants, is heard fairly often. It’s not clear exactly what that means. The goal of QE, and of monetary policy generally, is to set financial conditions consistent with full employment and stable prices, which can be thought of as trying to undo the economic distortions arising from price and wage stickiness, monopolistic competition, credit market frictions, and the like. In this respect appropriate monetary policy is “un-distorting;” in particular, allocations under active monetary policy should be closer rather than further from the competitive, free-market, flexible-price ideal. A possible rationalization of the “distortion” claim is that QE works, at least in part, by affecting term premiums (and through them, the whole gamut of asset prices). From a market participant’s point of view, when QE is active it feels as if government (central bank) decisions, rather than private-sector fundamentals, are setting asset prices. Moreover, in such situations it may appear that the highest returns go to the best Fed-watchers, rather than to those whose expertise is in evaluating economic fundamentals. Some frustration with this state of affairs on
the part of professional investors is understandable. Note, though, that QE affects term premium by affecting the net maturity distribution of government debt held by the private sector. In this respect, a QE program-which amounts to a replacement of longer-term government obligations in private hands with shorter-term obligations(bank reserves, in the case of QE)is not fundamentally different from a change in the maturity structure of debt issued by the Treasury That government decisions about the maturity structure of its debt would affect term premiums seems natural and, since the government has to choose some maturity distribution, it's not clear what it would mean for government policy to be " neutral " with respect to the term premium (Greenwood et al., 2014). In short, there is no such thing as an"undistorted" value of the term premium, not so long as the mix of outstanding government liabilities is relevant to asset pricing a possible response to this point is that at least Treasury maturity decisions are largely non-responsive to short-term economic conditions, with issuance policies generally being smooth and set well in advance. In contrast, Fed qe programs are typically large in size and less predictable, responding to economic developments and (importantly)to how monetary policymakers choose to interpret those developments. To the extent that Fed decisions are hard to forecast, even conditional on the outlook, they add noise to asset prices. But of course that is true for any form of monetary policy. I think it comes down to whether Fed policy, inclusive of policy errors and misjudgments, is economically stabilizing on net, or not. If it is stabilizing then though the unpredictable components of Fed policy and communication may be a nuisance for market participants, overall monetary policy(including QE)reduces rather than increases the overall level of distortions in the economy Another common critique of QE is that it purportedly promotes increased inequality, primarily because of its effects on the prices of stocks and other assets. This claim is questionable on its face(Bernanke, 2015; Bivens, 2015). Empirically, it is far from obvious that QE(or easy money generally) worsens inequality in any meaningful way, once all the diverse effects of policy are taken into account. It is of course true that, all else equal, higher stock prices mean greater inequality of wealth--although the effect on income inequality is mitigated by the fact that easy money also lowers the rate of return on assets, so that income from capital 8 As James Tobin(1977)once said, "It takes a heap of Harberger triangles to fill an Okun gap. "Translated from economy as a whole is operating far from its potential. The goal of monetary policy is to close OKlBpes hen the economist, this aphorism suggests that distortions at the microeconomic level are hardly of consequence w
13 the part of professional investors is understandable. Note, though, that QE affects term premiums by affecting the net maturity distribution of government debt held by the private sector. In this respect, a QE program—which amounts to a replacement of longer-term government obligations in private hands with shorter-term obligations (bank reserves, in the case of QE)—is not fundamentally different from a change in the maturity structure of debt issued by the Treasury. That government decisions about the maturity structure of its debt would affect term premiums seems natural and, since the government has to choose some maturity distribution, it’s not clear what it would mean for government policy to be “neutral” with respect to the term premium (Greenwood et al., 2014). In short, there is no such thing as an “undistorted” value of the term premium, not so long as the mix of outstanding government liabilities is relevant to asset pricing. A possible response to this point is that at least Treasury maturity decisions are largely non-responsive to short-term economic conditions, with issuance policies generally being smooth and set well in advance. In contrast, Fed QE programs are typically large in size and less predictable, responding to economic developments and (importantly) to how monetary policymakers choose to interpret those developments. To the extent that Fed decisions are hard to forecast, even conditional on the outlook, they add noise to asset prices. But of course that is true for any form of monetary policy. I think it comes down to whether Fed policy, inclusive of policy errors and misjudgments, is economically stabilizing on net, or not. If it is stabilizing, then though the unpredictable components of Fed policy and communication may be a nuisance for market participants, overall monetary policy (including QE) reduces rather than increases the overall level of distortions in the economy.8 Another common critique of QE is that it purportedly promotes increased inequality, primarily because of its effects on the prices of stocks and other assets. This claim is questionable on its face (Bernanke, 2015; Bivens, 2015). Empirically, it is far from obvious that QE (or easy money generally) worsens inequality in any meaningful way, once all the diverse effects of policy are taken into account. It is of course true that, all else equal, higher stock prices mean greater inequality of wealth—although the effect on income inequality is mitigated by the fact that easy money also lowers the rate of return on assets, so that income from capital 8 As James Tobin (1977) once said, “It takes a heap of Harberger triangles to fill an Okun gap.” Translated from economist, this aphorism suggests that distortions at the microeconomic level are hardly of consequence when the economy as a whole is operating far from its potential. The goal of monetary policy is to close Okun gaps
rises by less than the rise in asset values. However, QE also yields gains in income and wealth that are more broadly based, including 1)positive effects on house prices, the principal asset of the middle class; 2)the benefits of lower interest rates and higher prices for debtors, including homeowners able to refinance to lower payments; 3)the savings for taxpayers of lower government borrowing costs and(possibly) increased seigniorage, and 4 )most importantly, the effects of monetary accommodation on jobs, wages, and incomes (Bivens, 2015; Engen et al 2015). It's revealing that in public debates, advocates for workers--like the group Fed Up which met with FomC members in Jackson Hole in 2016--have tended to favor the continuation of easy money, while the typical op-ed about the adverse effects of easy money on the distribution of income and wealth is written by a hedge fund manager, banker, or right-wing politician--people who have otherwise not traditionally exhibited much concern about inequality (Fleming, 2016). That political alignment-workers' groups in support of easy money, financiers in favor of higher interest rates-is of course the historical pattern in the United States, going back to William Jennings Bryan and beyond In any case, whatever effects monetary policy has on inequality are likely to be transien in contrast to the secular forces of technology and globalization that have contributed to the multi-decade rise in inequality in the United States and some other advanced economies. If the monetary effects on inequality are modest(indeed, of indeterminate sign) and mostly temporary as seems most likely, then it makes sense for monetary policymakers to ignore distributional effects and to focus on their legal mandate to promote price stability and full employment leaving distributional concerns be addressed by other policies, including fiscal policy. If, on the other hand, the effects of monetary policy on inequality are not transient, then presumably the reason is what economists have called hysteresis, the idea that a" hot " economy promotes higher long-term growth by promoting labor force participation, higher skills, and higher wages However, the presence of significant hysteresis effects would likely imply that easy money during periods of economic weakness reduces inequality, rather than the reverse A final criticism of QE is that it exposes the central bank to capital losses, in the event that longer-term interest rates rise unexpectedly quickly. Although central banks dont have to mark to market, and they can operate perfectly well with negative capital, losses on their asset 9 Hence the apparently contradictory claims that QE both helps wealth-holders and hurts savers; see Bernanke (2015)for a discussion
14 rises by less than the rise in asset values.9 However, QE also yields gains in income and wealth that are more broadly based, including 1) positive effects on house prices, the principal asset of the middle class; 2) the benefits of lower interest rates and higher prices for debtors, including homeowners able to refinance to lower payments; 3) the savings for taxpayers of lower government borrowing costs and (possibly) increased seigniorage; and 4) most importantly, the effects of monetary accommodation on jobs, wages, and incomes (Bivens, 2015; Engen et al., 2015). It’s revealing that in public debates, advocates for workers—like the group Fed Up, which met with FOMC members in Jackson Hole in 2016—have tended to favor the continuation of easy money, while the typical op-ed about the adverse effects of easy money on the distribution of income and wealth is written by a hedge fund manager, banker, or right-wing politician—people who have otherwise not traditionally exhibited much concern about inequality (Fleming, 2016). That political alignment—workers’ groups in support of easy money, financiers in favor of higher interest rates—is of course the historical pattern in the United States, going back to William Jennings Bryan and beyond. In any case, whatever effects monetary policy has on inequality are likely to be transient, in contrast to the secular forces of technology and globalization that have contributed to the multi-decade rise in inequality in the United States and some other advanced economies. If the monetary effects on inequality are modest (indeed, of indeterminate sign) and mostly temporary, as seems most likely, then it makes sense for monetary policymakers to ignore distributional effects and to focus on their legal mandate to promote price stability and full employment, leaving distributional concerns be addressed by other policies, including fiscal policy. If, on the other hand, the effects of monetary policy on inequality are not transient, then presumably the reason is what economists have called hysteresis, the idea that a “hot” economy promotes higher long-term growth by promoting labor force participation, higher skills, and higher wages. However, the presence of significant hysteresis effects would likely imply that easy money during periods of economic weakness reduces inequality, rather than the reverse. A final criticism of QE is that it exposes the central bank to capital losses, in the event that longer-term interest rates rise unexpectedly quickly. Although central banks don’t have to mark to market, and they can operate perfectly well with negative capital, losses on their asset 9 Hence the apparently contradictory claims that QE both helps wealth-holders and hurts savers; see Bernanke (2015) for a discussion