Restarting quantitative easing (QE) is not on the policy agenda today. But sooner or later policy rates will fall back to the lower bound and central banks will have to decide whether to resume QE or not. Given mounting concerns about the cost involved (Chadha et al. 2021), that decision to relaunch QE cannot be taken for granted and will hinge on an assessment of the effectiveness of QE. That remains a matter of debate.
A huge amount has been written about how QE works (e.g. Den Haan 2016). That literature suggests three basic transmission channels. The first two – the portfolio balance and liquidity channels – explain how asset purchases can impact asset prices thanks to market segmentation and preferred habitats (Vayanos and Vila 2021) and improving market functioning (Gagnon et al. 2011), respectively. The third, the signalling channel, argues that QE can shift expectations of future monetary policy decisions, and thereby asset prices (Bauer and Rudebusch 2014).
Our goal is to predict whether central banks will resort to QE in future downturns, so it makes sense to focus on what central bankers say about QE. That is the purpose of this column, which focuses on one central bank (the Bank of England, or the Bank) and one key channel (signalling), drawing on the comments of policymakers and staff.
There is no groupthink about QE at the Bank. There are differences of opinion over time, as new evidence arrives and personalities change, and across individuals at a point in time. Nevertheless, it is fair to speak of a shift in emphasis in the Bank’s narrative on QE, away from the portfolio balance and liquidity channels towards signalling when markets are not dysfunctional (Broadbent 2018). However, the institutional set-up and the comments of policymakers raise questions about the Bank’s confidence in any of the mechanisms that might underpin that signal, and hence the significance of the channel itself. These concerns about the signalling channel are unlikely to be peculiar to the Bank.
Mechanism #1: Unconstrained at the lower bound
Bad things can happen when the policy rate reaches the lower bound in a world without unconventional monetary policy (and active fiscal policy). A constrained central bank cannot respond to negative shocks, and the economy can become trapped. A contraction in spending can trigger a decline in inflation expectations and rising real interest rates. The decision to launch QE can send an important signal in this scenario. By revealing that it is unconstrained, the central bank can reduce uncertainty and above all fears of very bad outcomes, and that could support spending (Gorodnichenko et al. 2022). The Bank believed in this mechanism when it began QE (Dale 2010) but there are three important caveats. First, the mechanism relies on the fact that QE ‘works’ through the other transmission channels (the central bank is not constrained). If asset purchases are ineffective then there is no reassuring signal. Second, the signal that the central bank is unconstrained and injecting effective stimulus should cause expectations of the policy rate in the future to rise, not fall (risk premia may compress on the good news, but bond prices likely fall). Third, this signal, and the associated shift in beliefs and spending, is only relevant at the launch of QE. Once unconventional policy becomes conventional nobody should need reassurance that the central bank is unconstrained. However, if policymakers publicly lose faith in the other transmission channels, then this mechanism could plausibly run in reverse: concerns about the central bank being constrained could start to re-emerge. Spending could stall.
Mechanism #2: Managing interest rate risk
The signalling mechanism favoured by many academics is the idea that the accumulation of financial assets leads to a shift in the reaction function towards ‘low for long’ as the central bank puts some weight on managing the interest rate risk in the QE portfolio (Bhattarrai et al. 2023). The mechanism is coherent – it explains how QE credibly signals a shift in the future policy path – but it is unclear whether any real-world policymakers take these considerations into account when they set monetary policy. This mechanism doesn’t apply to the Bank given that the QE portfolio is fully indemnified by HM Treasury (Ramsden 2023). Nor is it clear that this mechanism works as intended if policymakers buy very long-term bonds. Committing to a low for too long strategy might drag down forward rates at shorter horizons, but the impact on the price of, say, a 20-year bond might be more than offset by an increase in rate expectations at longer horizons and an increase in the term premium implicit in bond prices.
Mechanism #3: Mimicking a commitment
The Bank’s Governor (and co-authors) observed that “while the precise mechanisms for the signalling channel are less well established in the theoretical literature, the explanation tends to centre on QE as a way to mimic a commitment by the policymaker to hold rates lower for longer than they otherwise would” (Bailey et al. 2020). The low for too long strategy at the lower bound has an impeccable pedigree (Eggertsson and Woodford 2003). However, this ‘mimicking a commitment’ approach would presumably be more effective if it was “accompanied by explicit forward guidance, rather than regarded as a substitute for it” (Woodford 2012) and ideally, we would know the mechanism – the bees wax and rope – through which QE lends additional credibility, convincing investors that policymakers will stick to the strategy. Going forward, investors may be more sceptical on this point given the path of policy rates following the latest round of QE (Schnabel 2024). A former Bank Deputy Governor has retrospectively confirmed that the Bank has not attempted to send such a signal and argued there is no credible bees wax and rope, particularly when policymakers are trying to bind their successors (Broadbent 2022a). Joyce et al. (2011) confirm that the Bank did not use QE explicitly to “signal future intentions” about the path of policy at the outset.
Mechanism #4: Releasing private information
An alternative view of the QE signal is that it reveals private information held by the central bank to the market. For example, Vlieghe (2016) argues that QE might have sent a signal that the economy was weaker than previously thought or that low interest rates were less stimulatory than previously thought (because the neutral rate was lower). Once again, policymakers could communicate this private information explicitly (e.g. publish an estimate of neutral) rather than send an implicit signal via QE. Moreover, one would have to believe that there is a valuable stock of private information waiting to be disclosed every time a central bank engages in QE if this signalling channel is a standard part of the transmission mechanism. In contrast, a former Deputy Governor has argued that the Bank does not have any additional private information to reveal, and it would be very hard to agree any message to send (Broadbent 2022b). Nor is it clear that the release of such private information would stimulate the economy. The revelation that the economy is weaker than previously thought might cause expectations of the policy rate to fall, but it might also encourage people to save rather than spend. There is likely very little to be gained through this channel if market participants already expect the policy rate to remain at the lower bound for many years to come (Vlieghe 2019).
Mechanism #5: Purchases delay hikes
Finally, a more pragmatic interpretation of the QE signal is that it is a form of commitment about the near-term path of the policy rate. Investors may believe that a central bank would not raise the policy rate whilst asset purchases are ongoing and therefore a long QE programme might push back expectations of the timing of the first hike (Busetto et al. 2022). However, investors may simply believe that the policy rate will jump up at the end of the programme so the net impact of the signal on financial conditions is likely limited. Nor is it clear how and to what extent bond purchases lend additional credibility (have additional traction on the expected policy path) to orthodox communication about the timing of the rate hike – not least since MPC members have voted to raise rates during the most recent QE programme (BoE 2021).
Conclusions
The policy pendulum will swing back towards an ultra-accommodative stance at some point. Policymakers will have to choose between restarting QE, cutting into negative territory, or communicating credibly about the future rate path. But questions remain about the effectiveness of QE and hence whether the benefits of more asset purchases outweigh the costs. The signalling channel of QE is well established in the literature, but policymakers appear to have reservations about any of the underlying mechanisms that support that channel. The case for further QE could therefore hinge on the perceived power of the portfolio balance and liquidity channels, which have also been queried outside moments of market dysfunction. The return of QE in the next recession should not be taken for granted.
References
Bailey, A, J Bridges, R Harrison, J Jones and A Mankodi (2020), “The central bank balance sheet as a policy tool: past, present and future”, Bank of England Staff Working Paper 899.
Bank of England (2021), “Minutes of the November 2021 Monetary Policy Committee meeting”, 4 November.
Bauer, M and G Rudebusch (2014), “The signaling channel for Federal Reserve bond purchases”, International Journal of Central Banking 10(3): 233-289.
Broadbent, B (2018), “The history and future of QE”, Speech, 23 July.
Broadbent, B (2022a), “Reliable partners”, Speech, 30 March.
Broadbent, B (2022b), “Q&A at The MPC at 25 conference”, 30 March.
Busetto, F, M Chavaz, M Froemel, M Joyce, I Kaminska and J Worlidge (2022), “QE at the Bank of England: a perspective on its functioning and effectiveness”, Quarterly Bulletin, Q1.
Chadha, J, P Turner and W Allen (2021), “Quantitative tightening: Protecting monetary policy from fiscal encroachment”, VoxEU.org, 23 October.
Dale, S (2010), “QE – one year on”, Speech, 12 March.
Den Haan, W (ed.) (2016), Quantitative Easing, CEPR Press.
Eggertsson, G and M Woodford (2003), “The Zero Bound on Interest Rates and Optimal Monetary Policy”, Brookings Papers on Economic Activity 34(1): 139-211.
Gagnon, J, M Raskin, J Remache and B Sack (2011), “Large-Scale Asset Purchases by the Federal Reserve: Did They Work?”, International Journal of Central Banking 7(1): 3-43.
Gorodnichenko, Y, S Kumar and O Coibion (2022), “The effect of macroeconomic uncertainty on firm decisions”, VoxEU.org, 19 September.
Joyce, M, M Tong and R Woods (2011), “The United Kingdom’s quantitative easing policy: design, operation and impact”, Quarterly Bulletin, Q3, pp. 200–12.
Ramsden, D (2023), “Quantitative tightening: the story so far”, Speech, 19 July.
Schnabel, I (2024), “The benefits and costs of asset purchases”, Speech, 28 May.
Tenreyro, S (2023), “Quantitative easing and quantitative tightening”, Speech, 4 April.
Vayanos, D and J-L Vila (2021), “A Preferred-Habitat Model of the Term Structure of Interest Rates”, Econometrica 89(1): 77-112.
Vlieghe, G (2016), “Monetary policy expectations and long term interest rates”, Speech, 19 May.
Vlieghe, G (2019), “Monetary policy: adapting to a changed world”, Speech, 15 October.
Woodford, M (2012), “Methods of Policy Accommodation at the Interest-Rate Lower Bound”, paper presented at the Jackson Hole symposium on The Changing Policy Landscape, 31 August.
