Comparison of QE program design

Table 4 highlights some interesting differences between the Fed’s and the ECB’s QE program design.

Initially, the ECB’s combined purchases made until September 2016 should amount €1.14 trillion, which represents 11% of Eurozone GDP (Micossi, 2015). With the extension to March 2017 at the current rate of 80 billion the total size of the purchases would be €1.74 trillion (17% of GDP)(Marsh, 2016). The decision to reinvest the principal payments of maturing securities will add €680 billion by 2019, which is around 6.5% of the Euro area GDP (ECB, 2015a).

In contrast, after six years of QE, the total purchases of the US amounted to $3.7 trillion, which represents 22.5% of US GDP. However, the first and largest asset purchase program by the Fed, QE1, had a comparable size of 12% of US GDP at the beginning of the program. Still, the euro-area program is the smallest QE program so far with respect to the BOE QE program with a size of £375 bn. (22% of GDP) and the recent BOJ QE program with ¥130 tr. (27% of GDP) (Gros et al., 2015).

61

Table 4 Comparison of QE program design

European Central Bank Federal Reserve

Area Euro Area United States

Program Public Sector purchase program (PSPP) Large-Scale asset purchases (LSAPs) Period March 2015 – March 2017 November 2008 – October 2014

Main objective Price stability (2% inflation) Support economic recovery, dual mandate Communication/

transparency Immediate press conference, monthly

bulletin, annual reports FOMC immediate announcement, hearing before Congress, monthly reports

Forward guidance Threshold-based, Qualitative Calendar-based, Threshold-based Transmission

Channels

Portfolio rebalancing, signaling channel, bank lending, exchange rate

Portfolio rebalancing, signaling channel Types of assets Euro-area public and private debt

securities US government, MBS, and housing-related

securities issued or guaranteed by GSEs Strategy/Quantity Open-ended Fixed quantity (QE1, QE2), Open-ended (QE3)

Size EUR 1.74 tr. 17% of GDP USD 3.7 tr. 22.5% of GDP

Main policy rates

at start Main refinancing operations 0.05%

Deposit rate -0.20% Federal funds rate 1.00%

Deposit rate -0.40% 0.25-0.5%

Source: Fed & ECB

Figure 11 shows the balance sheet expansions of the ECB versus the Fed. Clearly, while the Fed’s balance sheet increased significantly after the financial crisis and even beyond its tapering in 2013, the ECB’s balance sheet actually declined between 2013 and 2015 until the start of QE.

Figure 11 Balance sheet ECB and Fed 2008-2016 (dollar/euro trillions)

Source: US. Bureau of Economic Analysis (2016), Statistical Data Warehouse (2016)

The Fed’s QE programs also differed in their strategies; QE1 and QE2 were fixed in their quantities

0 0.5 1 1.52 2.5 3 3.54 4.5 5

2008 2010 2012 2014 2016

ECB Fed

62

along the yield curve, Operation Twist had fixed quantities but at selected maturities with the aim of prolonging the average maturity of the Fed’s Treasury securities, and QE3 was open-ended and closely tied to policy goals (Ubide, 2014). The ECB’s APP, similar to the Fed’s QE3, is open-ended with its end-date and quantity linked to inflation goals.

The Fed’s and ECB’s programs also differ in the type of securities that the central banks purchase for open market accounts. The Fed was limited to assets that had the full backing of the US government, which is why the purchases consisted of US Treasury securities, housing related

government-sponsored enterprise debt, and mortgage-backed securities (MBS) guaranteed by the government.

The choice of securities also reflected the nature of the US financial system, where mortgages and the housing market play a crucial role in the transmission mechanism (Rosengren, 2015).

The PSPP includes nominal and inflation-linked central government bonds and bonds issued by recognized agencies, international organizations and multilateral development banks. There are basically no restrictions on the type of bonds the ECB can purchase as long as they are rated

investment grade. The program thus far included covered bonds, asset-backed securities, agency, and sovereign debt. The maturity restriction implies that only securities with a minimum maturity of two years and a maximum maturity of 31 years can be purchased (Benoît Cœuré, 2015). The Fed, on the other hand, had mostly purchased long-term securities, which means that unwinding the excess reserves will take considerable time. Further, by limiting the program to two years, or at least until policy goals are met, the ECB will in my view be able to more easily communicate an exit strategy.

The original program design quickly constrained the ECB asset purchases in countries with little public debt or no eligible agencies because of the issuer and issue share limits. Furthermore, as the

allocation of purchases is based on the countries’ capital key, this automatically allocated most purchases to German and French bonds, whose interest rates are already extremely low. Claeys et al.

(2015, 2016) showed, that the imposed limits would largely constrain the size and duration of the program, and thus potentially limit its effectiveness. With the latest changes to its program, the ECB greatly expanded the list of available debt securities for purchase. The ECB added regional and local Euro area government bonds and corporate debt to the list of eligible assets. The ECB limited itself to

63

purchases in the secondary market since the treaty restricts primary market purchases under the PSPP. However, corporate debt securities form an exception in order to fulfill liquidity needs to meet its targets (ECB, 2016f). Corporate bond purchases on a large scale are a new territory for the ECB and were not part of the Fed’s QE either. Adding corporate bonds to the purchase program supports the monetary transmission to the real economy by helping to directly lower borrowing costs for

businesses (ECB, 2016a). In my opinion, purchasing a broad basket of asset classes likely affects several transmission channels, which could facilitate spillovers to other assets and increase the chances for the ECB to meet inflations targets.

In the US, the main channels of transmission through which QE operated were the portfolio balance channel and the signaling channel. This has been confirmed by many empirical studies, while the relative importance of each transmission mechanism is less clear (Gagnon et al., 2011; Hamilton &

Wu, 2011; Krishnamurthy & Vissing-Jorgensen, 2011). In the Eurozone, other important channels through which QE can be expected to operate are the credit channel, especially bank lending, and the exchange rate channel. Even though the ECB denies directly targeting the exchange rate, a

depreciation of the Euro is highly favorable for trade balances.

Figure 12 depicts the USD/EUR exchange rate and shows the sharp depreciation of the euro against the US dollar following the QE announcement in late 2014 and early 2015. In search for low yields, rising demand for European bonds should have facilitated this trend. Preliminary evidence on the transmission of the ECB QE program has shown that because of the widespread asset composition, asset purchases have facilitated the duration channel, credit channel, and exchange rate channel and led to spillovers to various other assets (Altavilla et al., 2015). However, because of ultra-low interest rates in many European countries, the portfolio balance channel could be weaker than in the US.

Furthermore, ECB QE measures are largely designed to boost bank lending and restoring confidence by signaling its accommodative stance through sizable asset purchases, which can trigger effects through the bank lending channel. Evidently, there have already been improvements in financing

64

conditions¹, while effects of the new TLRTO’s and corporate bond purchases could further support transmission via the credit channel to the real economy.

Figure 12 USD/EUR foreign exchange rate

Source: US. Bureau of Economic Analysis (2016)

The ECB was the first central bank to implement negative deposit rates while conducting QE, meaning that the ECB is charging banks to hold excess reserves to encourage them to withdraw these deposits and stimulate lending to the economy. This stands in stark contrast with the Fed, which paid banks interest on reserves, thereby allowing the Fed to expand its balance sheet and still maintain control over the federal funds rate. Paying interest on excess reserves puts a floor on the federal funds rate since for banks, lending at rates below the interest rate on excess reserves is not attractive. By controlling the rate on reserves, the Fed can further change the incentives for depository institutions to hold reserves to a level in alignment with its monetary policy, which was crucial for the exit of QE (Econ, 2013).

In the Euro area, the negative rate on excess reserves does encourage a reduction of those deposits held at the ECB and thus would imply a shrinkage in the size of the balance sheet. One might argue that this conflicts the aim of QE as its intention is an expansion of the balance sheet, however, bank lending to the real economy remains a major concern for the ECB. The impact of the ECBs policies on inflation will ultimately depend on whether banks use the additional holdings of reserves for new

1 See section 4.3.5.

QE starts

More QE

More QE 1.00

1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40

Jun-14 Jan-15 Aug-15 Feb-16

65

lending. While the payment of interest on reserves by the Fed has led banks to hoard reserves, the ECB, by discouraging cash holding, could be more successful in stimulating borrowing and spending (Duprat, 2015b).

While the Fed has a dual mandate for maximum employment, stable prices, and moderate long-term interest rates, the ECB has a single mandate for price stability, which implies a target inflation rate below but close to 2%. As stated in Article 127(1) TFEU, the ECB is usually not concerned with economic recovery, only with price stability. The only reason, the ECB was able to implement QE measures despite large opposition was because deflationary pressures were severe enough to threaten its policy mandate (Pacces & Repasi, 2015).

It could be argued that the dual mandate was a defining motive for the Fed to move more

aggressively in its policies than any other central bank after the financial crisis. Inflation rates and their severance and persistence depend largely on the accuracy of forecasting and can be influenced by temporary supply shocks, as can be observed in the recent oil price developments that drag down on inflation worldwide. But in the light of low inflation expectations in combination with significantly high unemployment rates and large output gaps, the dual mandate of the Fed forced them to take immediate action. A delay in reacting to disinflation can have severe consequences for an economy, as over time they feed into the public expectations of very low inflation rates (Rosengren, 2015). It has therefore often been argued that the implementation of QE in Europe was ‘too little too late’ to stabilize inflation expectations. Furthermore, its vague definition of price stability creates a challenge in communicating the means within its policy mandate (Ubide, 2014). The single mandate with a focus on price stability, together with the treaty restrictions explain the ECB’s conservative approach to QE, but this might in my view also lessen the effect it has on inflation.

Forward guidance played a major role in supporting the Fed’s monetary policy at the zero lower bound. The FOMC’s first policy announcements started with a vague language but became more decisive after implementing a calendar-based forward guidance in 2011. Its statement to cut rates while warranting the low range “…at least through mid-2013” had large effects on the current and expected future interest rates (Williams, 2012). It later shifted towards a threshold-based forward

66

guidance with QE3, communicating that the low federal funds rate would remain until unemployment rates were improved. The threshold-based forward guidance implied a credible commitment to the policy goal. However, the policy also increased the complexity of the statement language which could have reduced its impact if it was not understood by the public (Dudley, 2013).

The ECB introduced forward guidance much later in 2013 when it announced that key interest rates would remain at low levels for an extended period. It has, however, also been more explicit in its communication from the very beginning, using threshold-based and qualitative forward guidance with a flexible time horizon, complemented with a description of macroeconomic conditions. The ECB started forward guidance before the room for further reductions in key policy rates were exhausted, which stands in contrast to the Fed, which used forward guidance only at the effective lower bound (ECB, 2014). They also linked their forward guidance on their APP to their policy objective from the start of the program, reinforcing the signaling channel. Furthermore, the ECB provided stronger interest rate guidance than the Fed in its latest announcement that policy rates are expected to remain at present or lower levels “for an extended period of time, and well past the horizon of our net asset purchases.” (Draghi & Constâncio, 2016).