Early EM rate hikes supporting EM currencies

Following key central bank policy meetings last week in Australia, the US and UK, short-end interest rate markets have turned more dovish. We had argued back on 2nd November that “hawkish interest rate markets may have got slightly ahead of themselves.”

The Federal Reserve and certainly the RBA cooled market expectations of rate hikes while the Bank of England flat-footed markets by leaving its policy rate unchanged at a record-low 0.10%. This was in line with our view that the Fed would “once again put some distance between the start of the QE taper and the start of the rate hiking cycle.

Within developed economies only the RBNZ and Bank of Canada have so far ended their asset purchases and only the Norges Bank and RBNZ have hiked their policy rates (by 25bp each) since the pandemic struck 20 months ago. In comparison 10 out of 21 major EM central banks have hiked their policy rate at least once in the past six months.

The spread between the EM and developed economy central bank policy rate, which had narrowed from about 413bp in March 2020 to 360bp last summer, has thus since widened to a 3-year high of 423bp. The pattern was similar following the Great Financial Crisis.

Monetary policy tightening is proceeding more slowly in developed economies despite headline CPI-inflation rising to a 13-year high of 3.9% yoy in September while in EM it was stable at 3.3% yoy. EM central banks, at least as a whole, have got a handle on inflation.

As a result in developed economies the central bank “real” policy rate collapsed to a multi-decade low of -4.0% at end-September, compared to +0.8% in EM. Excluding India and China the EM central bank real policy rate was still -2.0%, or twice as large as the GDP-weighted real rate in developed economies.

This divergence in real policy rates partly explains why EM currencies have outperformed developed currencies since late-April, in our view.

If the gap between developed and EM CPI-inflation continues to widen we think developed interest rate markets may once again find it difficult to buy into these central banks’ ultimately dovish thinking and short-end government yields may start to rise again.

 

Interest rate markets turn more dovish in wake of key central bank policy meetings

Throughout October 2-year government bond yields in developed economies continued to rise as hawkish interest rate market upped their expectations that central banks would tighten monetary policy in the face of high headline CPI-inflation. A hawkish turn by the Bank of Canada at its 27th October policy meeting and hawkish comments by Bank of England Governor Bailey in recent weeks arguably added to the upward momentum in short-end government bond yields.

However, following key central bank policy meetings last week in Australia, the United States and United Kingdom 2-year government bond yields have been broadly unchanged in the United States and Japan, fallen in Canada, Germany and New Zealand and collapsed in Australia and the United Kingdom (see Figure 1).

 

 

The result is that the GDP-weighted average of 2-year government bond yields in the United States, Japan, Germany, United Kingdom, Australia and Canada fell about 4bp between 2nd and 8th November to about 0.26% – within touching distance of the 3-week low of 0.24% recorded on 5th November according to our calculations (see Figure 2).

 

 

This is in line with our view that “hawkish interest rate markets may have got slightly ahead of themselves” (see Hawkish rates markets ahead of themselves, 2nd November 2021). Our argument was that “the surge in short-end government bond yields in most developed economies since early October has de-facto resulted in a tightening of monetary conditions […]. Effectively, markets have already done some of the central banks’ “dirty work”. The rise in short-end government bond yields reduces the odds of central banks hiking rates as aggressively as currently priced in by markets, in our view – a self-defeating prophecy (or prophet’s dilemma), at least partly. We think that in coming months some developed central banks could conceivably try to cool market expectations of rate hikes and more broadly monetary policy tightening in order to cap short-end government yields and avoid a rapid tightening of monetary conditions weighing on economic growth”.

We also argued that while US Personal Consumption Expenditure (PCE) had in recent months been running slightly above its long-term trend, it had not come close to making up for a “pandemic shortfall” which we estimate at $1.1trn. Moreover, the slowdown in PCE growth in Q3 helped cool inflationary pressures, which nevertheless remained elevated mainly because of supply-push inflation (including high commodity and energy prices) over which the Federal Reserve has arguably limited influence.

In the United Kingdom, we argued that the “economy faces a number of headwinds, namely i) acute labour shortages and supply-chain constraints and bottlenecks; and ii) multiple challenges to domestic demand. On the demand-front the issue is not weak household purchasing power or strained corporate balance sheets but rather tepid consumer and business confidence as a result of planned tax increases in April 2022, furlough having ended on 30th September, fuel-shortages, rising CPI-inflation and likelihood that Bank of England will hike rates next year” (see Sterling – Fuel for thought, 1st October 2021).

We would argue that the Federal Reserve and certainly the Reserve Bank of Australia at their policy meetings last week cooled market expectations of rate hikes while the Bank of England flat-footed markets which had priced in a 15bp hike by leaving its policy rate unchanged at a record-low 0.10%.

Reserve Bank of Australia drops yield-control but that’s where its hawkishness ended

The Reserve Bank of Australia policy statement by Governor Lowe last Tuesday was hawkish at first glance but the devil was in the detail. The RBA formally abandoned its 0.1% yield target for the government’s key April 2024 bond. However, dropping the target had been expected given that 3-year yields had risen well above 0.1% in recent weeks. Moreover, the RBA said it would maintain weekly asset purchases of AUD 4bn until at least mid-February 2022 – so no tapering of QE for at least another three months. Finally, Governor Lowe implied that the RBA policy rate would likely remain unchanged at a record-low of 0.10% until at least late 2023 while in his post-meeting webinar stated that “it is still entirely possible that the cash rate will remain at its current level [of 0.10%] until 2024”.

Federal Reserve’s “dovish taper” keeps policy rate hikes at bay for now

The Federal Reserve announced at its policy meeting on Wednesday that it would start tapering its monthly asset purchases of Treasuries and MBS later in November and would further reduce purchases in December and following months, at a pace that currently implies QE will end in mid-2022. This was in line with our view in September that “the Fed will refrain from giving a more specific timeline for the start of its taper and do so only at its 3rd November meeting to retain some data-dependent policy flexibility” (see Powell Put in play but greater challenges ahead, 1st September 2021).

However, the Federal Reserve kept its policy rate on hold, in line with expectations, once again arguing that labor market conditions had not yet reached levels consistent with the Committee’s assessment of maximum employment. This was in line with our view that the Fed would at its meeting “once again put some distance between the start of the QE taper and the start of the rate hiking cycle” (see Hawkish rates markets ahead of themselves, 2nd November 2021).

Bank of England’s “dovish hold” and confused message flat-foots interest rate and FX markets

The big dovish surprise last week came from the Bank of England, which surprised markets by keeping its policy rate unchanged at 0.10% at its meeting on Thursday. Markets had priced in a 0.15bp rate hike to 0.25% (and a total of 85bp of hikes by end-2022), while analysts had been divided between a hike and hold. Only two of the nine Monetary Policy Council (MPC) members – Saunders and Ramsden, arguably the two most hawkish members – voted for a 0.15bp hike. That was a “dovish hold” in our view as a hawkish-sounding Governor Bailey in an interview on 9th October had stressed the need to prevent high inflation from becoming permanently embedded and had done nothing in recent weeks to dispel market expectations of rates hikes. Moreover, the central bank in its latest Quarterly Monetary Report now forecasts that CPI-inflation will peak at 5% yoy in April 2022.

The only hawkish twist – albeit a modest one – was that recently appointed MPC member Catherine Mann joined Saunders and Ramsden in voting for an immediate end to the Bank of England’s purchases of Gilts. In any case the central bank’s stock holding of Gilts should, at the current rate of increase, reach its target of £875bn in the second-half of December at which point the Bank of England is due to end QE. Bailey on Friday defended the Bank of England’s decision to keep rates on hold with some arguably odd arguments.

Markets are still pricing in a more than 50% chance of the Bank of England hiking its policy rate 15bp at its 16th December policy meeting. However, 2-year Gilt yields which collapsed 21bp on Thursday – the largest daily fall since March 2020 – have since fallen a further 7bp to about 0.42%, within touching distance of the one-month low of 0.41% recorded on 5th November (see Figure 1). Sterling depreciated 1.1% in trade-weighted terms on Thursday – its largest daily fall since September 2020 – and has since rebounded only 0.3%. We are sticking with our view that the Bank of England policy rate will be below 1.00% by end-2022.

In summary only the Reserve Bank of New Zealand and Bank of Canada have so far ended their asset purchases and only the Norges Bank and Reserve Bank of New Zealand have hiked their policy rates (by 25bp each) since the pandemic struck 20 months ago (see Figure 3).

 

 

EM central banks normalising policy rate faster… as was the case in wake of GFC

The bottom line is that monetary policy tightening in developed economies is proceeding at a very slow pace, both in absolute terms and relative to Emerging Market central banks, despite the fact that in September headline CPI-inflation was higher in developed economies than in EM. This is a case of history repeating itself.

 

 

Central banks in developed economies front-loaded rate cuts at the beginning of the pandemic – cutting by 78bp in GDP-weighted terms in March 2020 alone – but have since kept the policy rate broadly unchanged around -0.09% or -9bp (see Figure 4)[1]. In comparison central banks in EM economies cut their policy rate more gradually – by about 80bp in March-August 2020 to about 3.5% – but have since hiked their policy rate 65bp to a 20-month high of about 4.1%. Ten out of 21 major EM central banks have hiked their policy rate at least once in the past six months. As a result the spread between the EM and developed economy central bank policy rate narrowed from about 413bp in March 2020 to 360bp last summer but has since widened to a 3-year high of about 423bp in late-October according to our calculations (see Figure 5).

 

 

The pattern was similar during and after the Great Financial Crisis. From September 2008 to December 2009 – the GFC peak – EM central banks cut their policy rates more aggressively than developed central banks (by about 288bp versus 235bp) but started hiking their policy rates sooner (in mid-2011). As a result the spread between the EM and developed economy central bank policy rate narrowed from about 515bp in October 2008 to 442bp in early 2010 before widening to a multi-year high of 566bp in December 2011 (see Figure 5).

In the aftermath of the GFC central banks in developed economies justified low policy rates by (in part) pointing to low headline CPI-inflation (see Figure 6). However, in the decade to December 2019 the GDP-weighted central bank “real” policy rate – the nominal policy rate deflated by headline CPI-inflation (ex-post) – averaged only -1.1% (see Figure 7). The developed central bank real policy rate was in positive territory in only three out of 122 months (in 2015) and then only just. By comparison the GDP-weighted central bank real policy rate in EM economies averaged about +1.6% in that 10-year period.

 

 

Collapse in developed real central bank policy rate and EM currency outperformance

However, it is in the past six months that the gap between developed and EM central bank interest rate policy has become more startling.

In developed economies headline CPI-inflation has continued to rise – hitting a 13-year high of about 3.9% yoy in September – while (most) central banks have largely sat on their hands. As a result the real policy rate collapsed to a multi-decade low of -4.0% at end-September (see Figure 7). In September 2011 it had fallen to “only” -2.6%.

In comparison EM central banks have continued to hike their policy rates even as EM headline CPI-inflation fell from 3.5% yoy in May to 3.3% yoy in August and September. As a result the EM real policy rate rebounded from +0.3% in May to +0.8% in September. One could argue that EM central banks, at least taken as a whole, have got a handle on CPI-inflation.

 

 

Of course these overall figures mask significant intra-country differences while at the same time reflecting relative GDP-weights.

At the one end of extremes the National Bank of Poland’s real policy rate was -5.8% at end-September (although it may have risen slightly in October following the NBP’s 40bp policy rate hike). At the other extreme, the People’s Bank of China’s real policy rate was +3.15% at end-September (using the 1-year Loan prime rate of 3.85% as the benchmark policy rate). The Reserve Bank of India also had a relatively high real policy rate of -0.35%. However, even if we exclude India and China – the two largest EM  economies – the EM real central bank policy was still -2.0% at end-September or twice as large as the GDP-weighted real rate in developed economies according to our calculations (see Figure 8).

 

 

The evolution of the central bank real policy rate (a crude proxy for inflation-adjusted local currency returns) – specifically the collapse in developed economies and relative stability in EM economies since April-2021 – explains in part why EM currencies have outperformed developed currencies, in our view. According to our calculations, a GDP-weighted basket of developed economy currencies has depreciated about 3.0% versus the US Dollar whereas a GDP-weighted basket of EM currencies has depreciated only 0.3% (see Figure 9). Even excluding the Chinese Renminbi – which has appreciated 1.3% versus the US Dollar since late-April – a GDP-weighted basket of EM currencies has depreciated only 1.7%.

 

 

Markets may once again struggle to buy into developed central bank dovishness

Developed central banks’ arguments for very low real policy rates have centred on two main planks, namely that monetary policy needs to remain stimulative to support labour markets in the wake of the pandemic collapse in demand and that the rise in headline CPI-inflation will in any case prove transitory. Put differently developed central banks on the whole think that real policy rates will “naturally normalise” medium-term and that in the meantime the bigger concern is still headline economic growth rather than (mostly) cost-push inflation.

That certainly appears to be the current thinking at the Federal Reserve, European Central Bank, Bank of Japan, Bank of England and Reserve Bank of Australia and it is a message which interest rate markets have taken on board since last week’s central bank policy meetings. However, if CPI-inflation continues to push higher – and the consensus forecast is that data out on 10th November will show that US headline CPI-inflation rose from 5.4% yoy in September to a 31-year high of 5.8% yoy in October – developed interest rate markets may once again find it difficult to buy into these central banks’ ultimately dovish thinking and short-end government yields may start to rise again, in our view.

 

[1] The Reserve Bank of Australia cut its policy rate 15bp to a record-low of 0.10% in November 2020, the Danmarks Nationalbank cut its policy rate by 10bp to -0.60% in September 2021, the Norges Bank hiked its policy rate 25bp to 0.25% in September 2021 and the Reserve Bank of New Zealand hiked its policy rate 25bp to 0.50% in October 2021.