Large output gaps look set to keep inflation low in most emerging markets over the next few years. But further out, we think that worrying public debt trajectories in some places (Brazil and South Africa), and greater emphasis on growth over inflation by central banks in others, could result in monetary policy being kept too loose for too long, creating medium-term inflation risks.
- Large output gaps look set to keep inflation low in most emerging markets over the next few years. But further out, we think that worrying public debt trajectories in some places (Brazil and South Africa), and greater emphasis on growth over inflation by central banks in others, could result in monetary policy being kept too loose for too long, creating medium-term inflation risks.
- Policymakers in most major EMs followed the DM playbook in March and April, implementing fiscal and monetary support packages, although these have varied in size and scope. Many EM central banks purchased government bonds to ease stress in financial markets at the peak of the crisis, while a few have gone further by delivering full-blown QE programmes to loosen monetary conditions.
- Spare capacity is likely to prevent policy loosening from pushing up inflation significantly in the next few years. But as output gaps close, inflation pressures could build, particularly if central banks let them. From an EM perspective, we see two main reasons why central banks might not step on the brakes in time.
- The first is that they’re focused on helping governments control large public debt burdens. This motive is particularly pertinent to Brazil and South Africa. But debt ratios have risen sharply across the board this year. So other central banks (notably those in Mexico and India) may also come under pressure keep real interest rates lower for longer.
- The second is that they’re focused on delivering stronger economic growth. With the recovery from this crisis likely to be slow, central banks could begin to attach more weight to economic growth and less to inflation. Some central banks in Central Europe already seem to fit into this category. But more countries could succumb to this amid a broader intellectual shift towards accepting higher inflation.
- For some EMs, the macro impact of higher inflation would be positive. In Korea, Taiwan and Thailand, headline rates are stuck at low levels (of 0-2%), nominal policy rates are close to zero and private sector deleveraging could continue to put downwards pressure on equilibrium real interest rates. Against this backdrop, a rise in inflation of a few percentage points would probably have more benefits than costs.
- But for other EMs, the costs of a rise in inflation might end up outweighing the benefits. This includes Brazil, South Africa, India and Mexico. While higher inflation might help to control debt burdens and boost near term growth, central banks in these countries run a greater risk of losing control of the process.
- The rise in inflation would be coming from a higher starting point, of around 3-6%. Long term inflation expectations are often above central bank targets and are not particularly well anchored. As such, headline inflation could rise more quickly, perhaps to the high single digits or low double digits.
- This is where the macroeconomic and financial market costs of high inflation become more pernicious. High inflation economies tend to suffer from large and abrupt falls in nominal exchange rates, higher real lending rates, and more widespread borrowing and saving in foreign currencies (risking balance sheet mismatches). Policymakers in high inflation economies also often experiment more with distortive policies like capital controls and multiple exchange rates, which weigh on growth and promote corruption.
Could the crisis lead to higher EM inflation?
We recently explained in a Global Economics Focus that, although low inflation is likely to be the story over the next couple of years, the huge amount of policy stimulus injected by DM policymakers could push up inflation further ahead. Among the major DMs, we think that these risks seem greatest in the US and perhaps the UK. This Focus digs into the medium-term inflation risks for emerging markets.
The first section sets the scene. The second section looks at which EMs may be vulnerable to higher inflation. The third section looks at the costs of higher inflation. We finish by fleshing out where inflation risks are to the downside.
Emerging market CBs turn unconventional
Policymakers in most major EMs followed the DM playbook in March and April, implementing fiscal and monetary support packages, although these have varied in size and scope. And while most EM central banks did not deliver full-blown QE programmes to loosen monetary conditions, they did purchase assets to boost financial market liquidity. This has caused money supply growth to pick up in many EMs, notably in Latin America. Chart 1 shows the rise in the narrow measure supply or M0 (i.e. cash in circulation plus commercial banks’ reserves), but measures of the broad money supply have been rising sharply too.
Chart 1: Money Supply (M0, Latest % y/y)
Sources: Refinitiv, Capital Economics
Low inflation to prevail over the next few years
While these measures have boosted growth in the money supply in many EMs, we still think that low inflation will prevail in most countries over the next few years.
One important point here is that the link between the money supply and inflation is not mechanical. Indeed, QE in developed economies following the 2008/09 Global Financial Crisis didn’t significantly boost inflation. This is mainly because the extra money got stuck in financial sectors. This time around, an important difference is that strong money supply growth has come alongside larger budget deficits. As a result, the increase in the supply of money (in both EMs and DMs) is likely to find its way to households and businesses.
Even so, it’s worth emphasising that households and businesses may stay cautious for some time once the virus has been controlled, for fear of second waves or another pandemic. As a result, they may wish to hold excess money balances for some time. And to the extent that households and businesses do spend this money, there is likely to be enough spare capacity in most countries to prevent this from significantly pushing up inflation in the near term.
Once spare capacity is used up, the extra demand could in theory boost inflation. As it happens, that process is not automatic either. It would depend on the strength of the link between output gaps and inflation (known as the Phillips Curve in the economics jargon). It is tricky to measure, particularly in EMs, and has probably weakened over the past decade as it has in DMs. Large informal sectors in many EMs (most notably, India, Mexico, South Africa and Peru) might also reduce unionisation and workers’ bargaining power, weakening the responsiveness of wages to inflation.
Central banks key to long run outlook for inflation
Over the long term, inflation pressures will build only if central banks let them. After all, the vast majority of EM central banks have now adopted inflation targeting frameworks, with price stability the key goal. The key questions are therefore a) how central banks will set policy in the long term and, b) related to this, what the broader goals of governments might be.
From an EM perspective, we see two main reasons why central banks might not step on the brakes in time.
1. Large public debt burdens
The first is that they are focused on helping to control large public debt burdens. Indeed, higher inflation can help to reduce real interest rates on government debt.
One of the great legacy costs of COVID-19 will be a sharp rise in public debt ratios. Aside from countries in acute debt crises (Argentina and a handful of lower-income EMs) these concerns are perhaps greatest in Brazil and South Africa.
Public debt trajectories in these two economies were concerning even before the crisis. Long term government bond yields in both countries are high, and real GDP growth is structurally weak. Public debt ratios will rise by a further 15-20%-pts this year (compared with 2019).
While both governments are planning austerity in the coming years, economic weakness is likely to weigh on political appetite for belt tightening before long. Indeed, even if governments unwind this year’s stimulus measures, they will still need to tighten fiscal policy aggressively to stabilise public debt ratios. (See Chart 2.)
Chart 2: Change in Primary Budget Balance Needed to Stabilise Public Debt Ratio by 2022 (% of GDP)
Sources: Refinitiv, IMF, Capital Economics
Against that backdrop, central banks could come under pressure to set policy with fiscal concerns in mind – a situation known as ‘fiscal dominance’. Keeping borrowing costs lower for longer could help governments to finance large deficits and reduce real yields.
Fiscal dominance comes under the umbrella of ‘financial repression’ – a term used to describe policies that keep borrowing costs artificially low. Indeed, we think that the governments of Brazil and South Africa are prime candidates to implement a broader set of financial repression policies to control debt ratios over the medium term. (See here.)
While fiscal positions are particularly dire in Brazil and South Africa, public debt ratios have risen sharply across the board this year. So other central banks may also come under pressure keep real interest rates lower for longer.
In India specifically, the public debt to GDP ratio is high too. And the government is already in a position where large primary deficits are being sustained only by historically strong nominal GDP growth. Elsewhere in Asia, concerns about fiscal dominance are brewing in Indonesia, with the central bank is part monetising the government’s deficit this year.
2. Central banks may put more weight on growth
The second reason is that central banks might be focused on delivering stronger economic growth.
The possibility of long-term scarring as a result of the pandemic is likely to increase the clamour for looser monetary (and fiscal) policy. And this could end up with central banks attaching more weight to growth and less to inflation.
There are already signs of EM central banks’ reaction functions shifting in this direction – most notably in Central and Eastern Europe (CEE, notably Poland and Hungary). The ties between central banks and governments have become closer since 2008/09. Central banks have pursued more dovish policies in line with governments’ staunchly pro-growth agendas. The key risk in CEE in particular is that monetary policy is not tightened fast enough as output gaps close, causing inflation to overshoot. After years of strong growth, capacity constraints were starting to kick in before the COVID-19 crisis. And much of the hit to labour markets is likely to be absorbed by generous government support schemes, which should pave the way for a solid rebound in demand over the next few years.
Elsewhere, pressure from the government in India may soon cause the RBI to become more growth focused. The government has replaced hawkish members of the board with doves. And the finance minister has often called for rate cuts. Similarly in Indonesia, a bill currently in Congress would expand its mandate to focus more on growth.
More countries could succumb to this amid a broader intellectual shift towards accepting higher inflation. The US Fed’s decision to adopt average inflation targeting could be the thin end of the wedge in terms of the global shift in thinking on the topic. Even if EM central banks do not follow the Fed’s lead and explicitly change their mandates, the greater acceptance for higher inflation among global policymakers could embolden those in EMs to be more tolerant of rising price pressures.
No one size fits all
Given the huge diversity within the emerging world, a rise in inflation of a few percentage points would have different macro impacts in different countries.
Overall, a small rise in inflation would probably be desirable in the likes of Korea, Taiwan and Thailand. This is for a few reasons. First, headline rates have been stuck at DM rates of 0-2% since 2015. (See Chart 3.) So a rise of a few percentage points would bring headline rates towards the 2-4% range that academics generally consider to be optimal, and reduce deflation risks.
Chart 3: Consumer Prices (% y/y)
Second and related, nominal policy rates are close to zero in these countries. And large private sector deleveraging could continue to put downwards pressure on equilibrium real interest rates. So higher inflation would give central banks more policy space by raising equilibrium real interest rates.
That said, in many other emerging economies, higher inflation would be more concerning. For a start, inflation could rise further and become more difficult to control. In most EMs, inflation tends to be in the region of 3-6% already, significantly higher than the DM average of 0.5%. (See Charts 4 & 5.)
Chart 4: Consumer Prices (% y/y)
Sources: Refinitiv, Capital Economics
Chart 5: Consumer Prices (Latest, % y/y)
Sources: Refinitiv, Capital Economics
What’s more, inflation expectations in many EMs are generally higher than in DMs and the low-inflation EMs mentioned earlier. Long term breakeven rates are running at around 4-6% in the likes of Brazil, South Africa and Mexico (see Chart 6 again), above central banks’ long-term targets. And inflation expectations in these countries are not particularly well anchored either – they often move with contemporaneous variables such as current inflation and currencies. This suggests that higher inflation could filter through into higher expectations, feeding back into higher actual inflation. The net result is that central banks in these countries run a greater risk of losing control of inflation. In this case, headline rates could soon drift into the high single digits or low double digits.
Chart 6: 10-year Breakeven Inflation Expectations (%)
Sources: Bloomberg, Capital Economics
Table : Costs & Benefits of Higher Inflation
S. Fischer (1993)
The relationship between inflation and growth is stronger at lower rates of inflation. When inflation is below 15%, a 1%-pt rise in inflation is associated with a 0.13%-pt decrease in GDP growth; when inflation is between 15% and 40%, a 1%-pt rise in inflation is associated with a 0.8%-pt decrease in GDP growth.
R. Judson & A. Orphanides (1996)
A reduction in the rate of inflation from 25% to 15%, is associated with an increase in GDP growth of around 0.5%-pts
M. Sarel (1996)
Finds the threshold at which the negative effects of high inflation will be felt is 8%. Argues that below this rate inflation may have no, or even a slightly positive, impact on growth. If the inflation rate doubles, then the GDP growth rate could fall by 1.7%-pts.
M. Bruno & W. Easterly (1997)
Find that the threshold for inflation above which it has a strong negative economic impact is 40%.
A. Ghosh & S. Phillips (1998)
Find that at inflation rates of 2-3%, there is a positive relationship between inflation and growth but above this inflation and growth are negatively correlated. Inflation rates above 20% are most often correlated with negative per capita growth rates.
M. Khan & A. Senhadji (2000)
There is a threshold for inflation, above which rises in inflation can have a strong negative impact on GDP growth. They find that this threshold is 1-3% for developed economies and 7-11% for emerging economies.
S. Kremer, A. Bick & D. Nautz (2011)
Find that the threshold inflation rate is 2.5% for developed economies, and 17.2% for emerging economies.
D. Baglan & E. Yolda (2014)
12% is threshold for the level of inflation that causes negative effects on growth. They also find that at very high levels of inflation, the relationship breaks down.
Costs incurred in “high inflation economies”
Inflation in the high single digits (or low double digits) is where the costs would become more significant. There are the often-cited microeconomic ‘menu’ and ‘shoe leather’ costs, which would kick in at these rates of inflation. There are some general macroeconomic costs too. It can obstruct the workings of the price mechanism – a change in relative prices can be more easily confused with inflation. That in turn can disrupt the optimal allocation of resources.
High inflation also creates uncertainty, reducing investment. The academic evidence on the impact of high inflation on GDP growth is mixed, but generally suggests a negative relationship. (See Table 1.)
More generally, there are four macroeconomic costs of higher inflation that are particularly pertinent to emerging markets.
First, high inflation economies tend to suffer large and often abrupt currency falls. Higher inflation differentials compared to a country’s trading partners, all else equal, causes its real exchange rate to appreciate. This means that its nominal exchange rate needs to weaken simply to maintain external competitiveness. As such, high inflation economies tend to suffer larger and, and often abrupt, nominal exchange rate falls. These large nominal exchange rate falls then further fuel inflation. Of course, currencies often diverge from their equilibrium levels for long periods, since nominal exchange rates in the short run are often driven by factors such as risk appetite. This helps to explain the tendency for real exchange rates to become overvalued in high inflation economies. That in turn often leads to large current account deficits, creating an over-reliance on capital inflows, and unbalanced growth.
Second, high inflation economies tend to have higher real lending rates. This is due to the risk premium associated with higher future inflation risks. The experience of Turkey suggests that real lending rates could rise quite sharply, depressing investment. This would be a particular concern for economies with already low investment rates, like Brazil, South Africa and Mexico. (See Chart 7.)
Chart 7: Total Investment Rate (2019, % of GDP)
Third, higher inflation can promote ‘dollarisation’. The practice of depositing funds and saving in foreign currency is still fairly common in many some EMs, as households and businesses see hard currencies as better stores of value. Indeed, in high inflation economies like Nigeria, Turkey and Lebanon, the share of foreign currency deposits in domestic banking systems is over 50%. (See Chart 8.)
Chart 8: FX Deposits in Domestic Banking Sector
Higher inflation could also encourage borrowing in foreign currencies too, since lending rates on FX loans tend to be lower. This could cause balance sheet mismatches as exchange rates weaken if households and firms do not have sufficient FX assets to cover their FX debts. Indeed, recall that large FX borrowing in private sectors was a key cause of EM banking crises in the 1980s and 1990s. Balance sheet mismatches could also flare up in the public sector too. In all, we estimate that total FX debts are largest in Chile, Argentina and Turkey. (See Chart 9.)
Chart 9: Total FX Debts (Ex. Fin Sec., Latest, % of GDP)
Sources: Refinitiv, National Central Banks, Capital Economics
Finally, high inflation can also lead to distortions in the form of multiple exchange rates and capital controls. (See Argentina, Venezuela, Nigeria, Egypt and Lebanon in recent years.) The main reason for this is that policymakers often try to resist the downwards pressure on nominal exchange rates that comes from higher inflation. This can lead to capital controls and multiple exchange rates if policymakers then limit access to foreign currency via certain rates.
Capital controls and multiple exchange rates can create huge distortions and opportunities for corruption, which all weigh on growth. And they only delay the inevitable currency adjustments required to maintain external competitiveness.
High inflation difficult to slay
All of this begs the question: if high inflation starts to have malign consequences, won’t policymakers just try to reduce it if and when it becomes a problem?
The problem is that once inflation is very high, it is difficult to get out of the system. inflation Politically and economically painful policies are required. History suggests that bringing inflation down into the single digits requires very high real interest rates for a prolonged period, tackling indexation, drastic improvements in central bank credibility and an end to deficit monetisation. Appetite for aggressive policy tightening was weak before the pandemic. But the prospect of long-term economic scarring caused by the virus could further reduce appetite for policy tightening over the long term. As a result, policymakers might end up reluctantly accepting higher inflation, viewing it as the ‘least bad option’.
Edward Glossop, Senior Emerging Markets Economist, +44 (0)7896064878, firstname.lastname@example.org