Powerless central banks driven to tiers

• With interest rates near the lower bound, central banks are experimenting with tiers.
• But they reduce the impact of rate cuts and do little to support banks’ profitability.
• Such efforts add to signs that monetary policy is near its limits.

Recent weeks and months have seen a more marked shift towards monetary policy loosening on a global scale. Most Emerging Markets, and to some extent the US, have scope to generate stimulus through interest rate cuts. But since the ECB, Bank of Japan and Swiss National Bank (SNB) already have negative policy rates, they have had to think more imaginatively. The ECB introduced a tiered interest rate system for the first time this month to allow it to cut its short-term policy rate without leaving banks worse off. The SNB altered its existing tiers to exempt a greater share of banks’ reserves from the negative rate. And there have been some suggestions that the Bank of Japan might soon follow by adjusting its own tiering system.

A key argument against negative interest rates has been that they are damaging to banks. Negative central bank deposit rates require commercial banks to pay a fee on their reserves, which they have typically been unable or unwilling to pass on to their own depositors. For example, Chart 1 shows how euro-zone retail deposit rates have held just above zero since the deposit rate went negative in 2014.

Chart 1: Euro-zone Retail Deposit Rate & ECB Deposit Rate (%)

Sources: Refinitiv, S&P

The cost of the negative rate has therefore been a direct squeeze on banks’ profits.

The solution adopted by the ECB, SNB and BoJ has been to exempt certain shares of reserves from the negative rate. The details are shown in Table 1.

Table 1: Details of Negative Policy Rates
Central bank and date rates turned negative	Current interest rate	Scope
ECB (Jun. 2014)	-0.5% (deposit rate)	Previously applied to all “excess” reserves (€1850bn or 5.7% of assets). Now applies only to reserves seven times larger than required (€1080bn or 3.3% of commercial bank assets).
Swiss National Bank (Jan. 2015)	-0.75% (policy rate)	Initially applied to reserves more than 20 times larger than required (CHF264bn, 7.6% of assets). Now applies only to reserves more than 25 times larger than required and the share that is “required” has also increased (CHF157bn, 4.5% of assets).
Bank of Japan (Jan. 2016)	-0.1% (interest on policy rate balance)	Different rates paid on 3 tiers of reserves, with negative rates on reserves beyond a high threshold amounting to only 5% of banks’ deposits (¥19tn or 1% of commercial bank assets). On net, BoJ still pays banks positive interest on reserves.
Sources: Central banks, Capital Economics

This reduces the strain on banks as the average interest on their central bank reserves rises as a result. But, in theory at least, it shouldn’t dissuade them from lending or buying assets. Since they typically still have significant reserves which are not exempt, it will be the deposit rate which is relevant for decisions at the margin.

This has given the central banks in question a little more room for manoeuvre by reducing opposition to negative rates from the banking sector and resistance from within their own policy-making committees. But we doubt that interest rate tiering is a game changer for three reasons.

First, the direct effect of negative interest rates on banks’ profitability has been small. Table 1 showed that they typically apply to only a small share of total commercial bank assets and the fee itself is less than 1% in every case. The oft-cited €7.5bn cost in the euro-zone case is equivalent to only 0.02% of banks’ assets. Banks’ profitability has been much more affected by structural and other factors, which tiering won’t help to address.

Second, the tiering of deposit rates might reduce their effectiveness. Banks may well be less likely to pass the cost of negative rates on to their customers if a tiered system is in place, helping to lower any political hurdles to more negative official interest rates. But this also means that the incentive for firms and households to spend will be lower than it would have been if the negative rate was passed on in full. And some press reports have argued that, despite the central banks’ hopes to the contrary, the higher average rate on banks’ reserves caused by tiering is dis-incentivising interbank lending. Chart 2 shows that 3-month interbank lending rates have nudged up since tiering was introduced and enhanced in the euro-zone and Switzerland respectively, although we suspect that this mainly reflected other factors.

Chart 2: Three-month Interbank Interest Rate (%)

Source: Refinitiv

Third, tiered interest rates do not address the bigger problem of low long-term yields and flat yield curves. These have done more damage to banks than negative deposit rates through limiting their ability to raise profits by borrowing over short horizons and lending for longer. Low returns have also caused serious problems for pension funds, which have struggled to generate the returns they need to meet their long-term liabilities. (See this Global Economics Focus.) If anything, the shift to more tiering has probably served to reinforce expectations that interest rates will stay low for a very long period.

We have argued that the Bank of Japan is unlikely to follow in the other banks’ footsteps by using tiering as a means to cut interest rates further. (See this Japan Economics Weekly.) This is partly for the three reasons outlined above. But the BoJ also has a lot less scope to exempt more reserves from its negative rate since it already applies it to just 5% of reserves. Rather than cutting interest rates when it next meets at the end of October, we see the BoJ increasing its ETF purchases and “twisting” its JGB purchases towards the shorter end of the curve. The purpose of the latter would be to create more of an upward slope to support the financial sector.

The US Fed is somewhat removed from this problem, with scope to cut interest rates by nearly 200bps before it hits zero. But we doubt that the possibility of tiering would be enough to encourage it to go negative, even if it were warranted by economic conditions. As we explained in a recent US Economics Update, there are a couple of quirks in the US financial system that could make implementing a negative policy rate very difficult. What’s more, unlike the ECB, it faces no political or legal limitations on the quantity of assets that it can buy or hold. It would probably therefore choose more quantitative easing over negative interest rates and their associated side effects.

The upshot is that tiered interest rates have given some central banks a little more scope to loosen policy. But the effects of the latest interest rate cut in the euro-zone and likely future cuts in Switzerland will be even smaller than the cuts that preceded them. And we still doubt that the Bank of Japan will reduce interest rates further, regardless of tiering. This latest foray further into uncharted territory has served mainly to highlight that central banks are reaching their limits. In the absence of a major fiscal boost, we expect growth and inflation in Europe and Japan to remain weak for some time to come.

Review of recent policy changes

Since our last Global Central Bank Watch, monetary policy has remained dovish. Of the 20 central banks covered in this publication, only the Norges Bank raised rates (by 25bp), while 10 others cut rates. Notable changes include the 25bp cut by the Fed, the ECB’s 10bp cut and announcement of an open-ended QE programme, the surprise 50bp cut by the Reserve Bank of New Zealand, the 35bp cut by the Reserve Bank of India, and the smaller-than-expected 5bp cut to the one-year Loan Prime Rate (LPR) by the People’s Bank of China.

Chart 3: Changes in Benchmark Rates

Sources: Bloomberg, Capital Economics

In other developments, the Fed made some minor policy tweaks to try and alleviate the liquidity problems that sent money market rates soaring this month. These included cutting both the interest on excess reserves (IOER) and the offer rate on the reverse repo facility by 30bp. And the ECB’s policy package included other small changes worth noting. First were favourable adjustments to the terms of loans under TLTRO-III which could provide further support for bank lending. Second was the introduction of a tiered interest rate aimed at reducing the cost of negative rates. However, the effects of these measures are likely to be small. Indeed, in the press conference accompanying the meeting, Mario Draghi revealed that the Council was unanimous that fiscal policy should become the “main tool” to support growth and inflation in the near term – implying that euro-zone monetary policy is more or less at its limits.

What’s next?

Among the advanced economies, we think that the Fed will reduce interest rates by another 25bp at its December meeting. This is broadly in line with market expectations for end-2019. However, unlike the futures market, we don’t expect more rate cuts next year. (See our US Economics Update.)

In the euro-zone, we have pencilled in another 30bps of cuts by the end of next year. This would take the deposit rate to -0.80%, closing in on the limit to how low interest rates can go. In addition, we expect the pace of asset purchases to be ramped up to €30bn per month from next July. (See our European Economics Update.) Incoming ECB President Christine Lagarde has signalled she intends to continue with Mr Draghi’s dovish approach, commenting that “highly accommodative monetary policy” was still necessary to respond to the euro-zone’s near-term challenges for a sustained period.

Elsewhere in Europe, we expect the Norges Bank to now keep rates on hold until 2022 after September’s hike. We are generally more dovish than the market on Denmark, Sweden and Switzerland, where we expect continued rate cuts into early 2020 as weak euro-zone growth weighs on inflation. (See Table 2.) The Bank of England will probably hold off on any rate hikes until late next year, though much will of course depend on the Brexit process. In Japan, further easing this year is likely to be limited to increased ETF purchases. And, in an attempt to steepen the sovereign yield curve without cutting its policy rate deeper into negative territory, the Bank may change its purchases to include more shorter-term bonds.

Table 2: Summary of CE Forecasts for Policy Rate Changes by the End of 2019
Policy Direction	Economies
Easing	US, Australia, Canada, China, India, Mexico, Turkey, South Korea, Euro-zone, Brazil, Sweden, Japan
No Change	UK, Denmark, Poland, Norway, Switzerland, South Africa, Russia, New Zealand
Tightening	None
Source: Capital Economics

Meanwhile, we think that interest rates have further to fall in Australia and New Zealand into 2020, where weak labour market data and sluggish GDP growth are weighing on inflation. Similarly, in Canada, we expect muted GDP growth this year to prompt the Bank to cut interest rates three times, starting in October.

Moving on to EMs, China’s PBOC disappointed expectations for a cut in the rates at which it lends to banks in the interbank market in September. This led to a rise in Chinese bond yields. What’s more, the ability of monetary easing to stimulate credit growth in China appears to have diminished. The PBOC is likely to respond in time by loosening policy by more than most anticipate. (See our China Economics Update.) We think the 7-day reverse repo rate will be reduced by 50bps to 2.05% by end-2019. And we have pencilled in a further 25bp cut alongside the renminbi falling to around 7.5/$ next year as the US-China trade war escalates.

Chart 4: Change to Policy Rates by End-19 Implied by OIS Markets & CE Forecasts (%)

Sources: Bloomberg, Capital Economics

In India, the larger-than-expected August rate cut by 35bp is likely to be followed by another 25bp cut in October. The loosening of monetary policy this year risks going too far and has been in part at the behest of the government, increasing concern about the bank’s independence. (See our India Economics Update.) Elsewhere, political pressure at Turkey’s central bank means it will probably fully reverse last year’s monetary tightening. We think that the one-week repo rate will be lowered by a further 400bps by early 2020, but renewed lira weakness will eventually force policymakers to change course.

Finally, the surprisingly dovish statement by Brazil’s policymakers led us to pencil in another 25bp cut in October, following the 50bp cut this month. However, the scope for policy loosening depends on the success of proposed reforms to the pension and tax systems, which may face political opposition. And in Mexico, a sharp drop in inflation in August paved the way for an aggressive easing cycle, with a probable 25bp cut in December followed by a further 150bps of cuts in 2020.

---

Table 3: Central Bank Policy Rates
Country	Policy rate	Latest	Last Change	Next Change (CE Forecast)	End-2019	End-2020	End-2021
Major Advanced Economies
US	Fed funds target	1.75-2.00	Down 25bp (Sep. 2019)	Down 25bp (Dec. 2019)	1.50-1.75	1.50-1.75	1.50-1.75
Euro-zone	Deposit rate	-0.50	Down 10bp (Sep. 2019)	Down 10bp (Mar. 2020)	-0.50	-0.80	-0.80
Japan	Interest on excess reserves	-0.10	Down 10bp (Jan. 2016)	None on horizon	-0.10	-0.10	-0.10
UK*	Bank Rate	0.75	Up 25bp (Aug. 2018)	Up 25bp (Nov. 2020)	0.75	1.00	1.25
Other Advanced Economies
Canada	Overnight target rate	1.75	Up 25bp (Oct. 2018)	Down 25bp (Oct. 2019)	1.25	1.00	1.00
Australia	Cash rate	1.00	Down 25bp (Jul. 2019)	Down 25bp (Nov. 2019)	0.75	0.50	0.50
Switzerland	Sight deposit rate	-0.75	Down 50bp (Jan. 2015)	Down 25bp (Q1 2020)	-0.75	-1.00	-1.00
Sweden	Repo rate	-0.25	Up 25bp (Dec. 2018)	Down 25bp (Dec. 2019)	-0.50	-0.50	-0.50
Denmark	Deposit rate	-0.75	Down 10bp (Sep. 2019)	Down 10bp (Mar. 2019)	-0.75	-1.05	-1.05
Norway	Sight deposit rate	1.50	Up 25bp (Sep. 2019)	None on horizon	1.50	1.50	1.50
New Zealand	Cash rate	1.00	Down 50bp (Aug. 2019)	Down 25bp (Feb. 2020)	1.00	0.75	0.75
Major Emerging Economies
China	7-day reverse repo rate	2.55	Up 5bp (Mar. 2018)	Down 25bp (H2 2019)	2.05	1.80	2.25
India	Repo rate	5.40	Down 35bp (Aug. 2019)	Down 25bp (Oct. 2019)	5.15	5.50	6.00
Brazil	Selic rate	5.50	Down 50bp (Sep. 2019)	Down 25bp (Oct. 2019)	5.25	5.25	5.25
Russia	1-week repo rate	7.00	Down 25bp (Sep. 2019)	Down 25bp (Oct. 2019)	6.75	6.50	6.50
Mexico	Overnight target rate	8.00	Down 25bp (Aug. 2019)	Down 25bp (Sep. 2019)	7.50	6.50	6.00
South Korea	Base rate	1.50	Down 25bp (Jul. 2018)	Down 25bp (H2 2019)	1.25	1.00	1.50
Turkey	1-week repo rate	16.50	Down 325bp (Sep. 2019)	Down 200bp (Oct. 2019)	13.50	14.00	17.50
Poland	Reference rate	1.50	Down 50bp (Mar. 2015)	None on horizon	1.50	1.50	1.50
South Africa	Repo rate	6.50	Down 25bp (Jul. 2019)	None on horizon	6.50	6.50	6.50
Sources: Bloomberg, Capital Economics. *Based on a scenario in which Brexit is repeatedly delayed. For forecasts based on a deal or a no deal, see our UK Economics Update “Pick your own Brexit forecast”, 1st July 2019.

Table 4: Quantitative Easing & Other Unconventional Policies
Central bank	Monetary Base* (% of GDP)		Planned Asset Purchases		CE Forecast of Future Changes
Federal Reserve Bank	15%		None on horizon.		The Fed ended the run-down of its balance sheet in August and will continue to roll off up to $20bn of MBS a month and purchase offsetting amounts of Treasury securities. This means that once again the Fed is a net buyer of Treasury securities. However, after the recent money market liquidity problems, the Fed may need to resume “organic growth” of its balance sheet sooner than expected.
European Central Bank	26%		Monthly pace of €20bn.		The ECB relaunched its QE programme in September 2019 at a monthly pace of €20bn, which we expect to increase to €30bn from mid-2020. The programme is open-ended which suggests that the ECB is locked into QE for several years. The new round of purchases will be “by and large the same kind of assets” as those bought previously, implying that around 75% will be public sector and the rest private sector securities. There is a 33% limit on the share of a country’s bonds it can hold.
Bank of Japan	92%		Annual pace of ¥80trn.		While the Policy Board has reiterated that it will continue purchasing JGBs at an annual pace of ¥80 trillion, the actual pace of purchases has been well below that recently. With bond redemptions rising, we think that the annual net increase in JGB holdings will fall below ¥20trn in 2020.
Bank of England	26%		None on horizon.		The BoE has said that it will maintain assets at their current level of £445bn until Bank Rate has reached a level from which it can be “materially” cut. We suspect that this means around 2%, which is at least a couple of years away. If the UK leaves the EU without a Withdrawal Agreement, the Bank may restart its QE programme and Funding for Lending Scheme.
Swiss National Bank	79%		Occasional FX intervention.		The SNB intervened in the FX market to weaken the franc in mid-2019 and will continue to intervene if needed to prevent the franc rising too far, particularly when QE resumes in the euro-zone.
Riksbank	10%		SEK 45bn of government bonds between July 2019 and December 2020.		The Riksbank announced at its April meeting that it will buy SEK 45bn of government bonds between July 2019 and December 2020. However, it will stop reinvesting the principal and coupon payments from government bonds bought under its QE programme from July. Our view is that the balance sheet will not shrink significantly until at least 2023.
Sources: Central banks, Capital Economics. *Latest
Table 5: Calendar of Policy Decisions
Date		Economy		Policy Instrument		Prior	Survey	CE Forecast
25th September		New Zealand		Cash rate		1.00	1.00	1.00
26th September		Mexico		Overnight target rate		8.00	7.75	7.75
1st October		Australia		Cash rate		1.00	1.00	0.75
2nd October		Poland		Reference rate		1.50	–	1.50
4th October		India		Repo rate		5.40	–	5.15
16th October		South Korea		Base rate		1.50	–	1.50
24th October		Euro-zone		Deposit rate		-0.50	–	-0.50
24th October		Sweden		Repo rate		-0.25	–	-0.25
24th October		Turkey		1-week repo rate		16.50	–	14.50
24th October		Norway		Sight deposit rate		1.50	–	1.50
25th October		Russia		1-week repo rate		7.00	–	6.75
30th October		United States		Fed Funds target range		1.75-2.00	1.75-2.00	1.75-2.00
30th October		Canada		Overnight target rate		1.75	–	1.50
30th October		Brazil		Selic rate		5.50	–	5.25
31st October		Japan		Interest on excess reserves		-0.10	–	-0.10
5th November		Australia		Cash rate		–	–	0.75
6th November		Poland		Reference rate		–	–	1.50
7th November		United Kingdom		Bank rate		0.75	–	0.75
13th November		New Zealand		Cash rate		–	–	1.00
14th November		Mexico		Overnight target rate		–	–	7.50
21st November		South Africa		South Africa		6.50	–	6.50
29th November		South Korea		Base rate		–	–	1.25
3rd December		Australia		Cash rate		–	–	0.75
4th December		Poland		Reference rate		–	–	1.50
4th December		Canada		Overnight target rate		–	–	1.25
5th December		India		Repo rate		–	–	5.15
11th December		United States		Fed Funds target range		–	1.50-1.75	1.50-1.75
11th December		Brazil		Selic rate		–	–	5.25
12th December		Euro-zone		Deposit rate		–	–	-0.50
12th December		Switzerland		Sight deposit rate		-0.75	–	-0.75
12th December		Turkey		1-week repo rate		–	–	13.50
13th December		Russia		1-week repo rate		–	–	6.75
19th December		Japan		Interest on excess reserves		–	–	-0.10
Sources: Bloomberg, Capital Economics

---

Jennifer McKeown, Head of Global Economics Service, +44 20 7811 3910, jennifer.mckeown@capitaleconomics.com
Bethany Beckett, Assistant Economist, +44 20 7808 4052, bethany.beckett@capitaleconomics.com