- Sir Humphrey Appleby, Yes Minister, the BBC, 1980.
The fictional Sir Humphrey may well be an economist trying to answer whether EM central banks should tighten monetary policy in response to the recent rise in food and oil prices. Even as recently as a couple of months ago, the reluctance of EM central banks to hike rates was easily reconciled with tenuous growth in developed economies and the expectation that food price inflation would subside after transitory supply shocks wore off.
A lot has changed since. Growth in the developed markets - led by the US - looks in much better shape. Global growth too looks better balanced as EM growth looks set to ease in some of the fast-growing economies there while the moderation in global imbalances continues through narrowing current accounts. Not all developments, however, have been positive. Food price inflation is grabbing headlines again, as is the price of oil during its steady march to a US$100 price tag. Under such circumstances, should EM central banks still continue to tighten policy only to bring growth back to trend, or should they respond to the shocks to food and oil prices also? We go one step further than Sir Humphrey and state that EM central banks will likely continue to focus on the growth picture because increases in food prices appear to be caused by supply factors they cannot affect, whereas the run-up in oil prices is more a symptom of the global recovery than a shock.
Dealing with Food Price Inflation
Should central banks respond to the recent food inflation? Food inflation has been high in most EM economies across the CEEMEA, AXJ and LatAm regions recently. Given the large weight that food inflation commands in the computation of CPI inflation, headline inflation has naturally ticked up as well. The most persistent rise in food prices has been in the AXJ region where food inflation has been consistently outstripping headline inflation, indicating more benign core inflation. The CEEMEA region has seen a sharp rise in food prices but it is more in line with headline inflation. This is the case in the LatAm region as well, where food inflation is rising but not yet well ahead of headline inflation. The weight of food inflation in the overall CPI basket is consistently high in the AXJ region, while the picture is extremely varied in the CEEMEA and LatAm theatres. Clearly, food inflation is an important component of CPI baskets in EM economies, and a closer look at food prices and the monetary policy reaction is certainly warranted.
Transitory supply shocks to relative prices - not fertile ground for monetary policy action: In the face of repeated shocks to food prices and therefore to headline inflation, should central banks not respond with some form of monetary tightening? The answer lies in the reason behind the inflation in food prices. Food prices have risen because of global supply shocks (as La Niña plays havoc with temperatures, rain and wind) as well as local ones. These supply shocks are transitory in nature and are clearly not something that central bankers can ease with tighter policy. Tighter monetary policy would not help for at least two reasons: i) it would not lead to a faster resolution of these supply problems; and ii) it is a blunt tool that is better equipped to deal with aggregate price shocks rather than relative price movements such as food price inflation.
A natural adjustment mechanism from the first principles of economics: Finally, the effect of higher food prices on household consumption will be to reduce the use of expensive ingredients to the extent possible (the substitution effect) but will also result in postponing consumption of other goods until food prices stabilise (the income effect). A temporary slowdown in consumption that attempts to reduce the demand for an expensive product is beyond the scope of monetary policy but quite naturally delivered by the rational behaviour of households.
When should central banks respond to higher food prices? Ignoring or looking through food inflation spikes is not always advisable. One clear case where fighting off food inflation makes sense is if there is a risk of contaminating ‘core' prices. This is most likely when the slack in the economy is dwindling or has already disappeared, putting overall prices under pressure to rise. Contamination of core prices (or ‘second-round effects') can occur in at least three ways: i) sustained food price inflation can lead to an increase in overall inflation expectations; ii) bargaining for higher nominal wages, thanks to a decline in real wages as a result of food inflation; and iii) the inability of economic agents to distinguish between relative price shocks and aggregate price shocks, resulting in all prices being marked up, generating inflation. Clearly, all three channels are more active when the overall level of activity in the economy is high and labour markets are tight.
The risk of a pass-through to core prices is already rising in some countries. In China, India and Indonesia, there are signs that overall inflation expectations are rising. In Russia and Poland, general inflation pressures are rising (albeit at much more benign levels in Poland), which makes the pass-through from food prices a matter of concern for the central banks there (again, with much less upside risk in Poland). Finally, in the LatAm region, Brazilian inflation due to strong domestic demand has been a concern, but we expect the recent uptick in inflation, largely from food inflation, to be short-lived.
Does ‘core' inflation make sense as a concept in the EM world? Where food prices account for a small portion of the CPI basket, the use of a ‘core' concept clearly makes sense since food prices are usually more volatile. However, in the EM world, food prices can account from 14% (in South Africa) to 50.3% (in Ukraine) of the CPI basket. Under these circumstances, does the concept of ‘core' even apply? The short answer is yes. Trying to dampen volatility in headline inflation in order to ‘protect' the consumer may actually lead to a more volatile policy path and therefore to a more volatile outlook for growth, in our opinion. Eventually, this might do more harm than good from the economy's and therefore the consumer's point of view. Put another way, if food inflation is transitory and driven by supply shocks, then reacting to it means a very short horizon for the inflation target. Choosing to look at core prices minimises the risk of policy over-reaction and likely protects the interests of the economy and consumers. In fact, the discussion above about food price inflation serves to illustrate this point rather well. Of course, the risk is that central banks may rely too much on these shocks unraveling without spreading to core prices, which is a risk that is rising in many emerging economies as discussed above.
Dealing with Oil Price Rises
Oil prices: Endogenous: The last time oil prices breached US$100, a significant amount of real and virtual ink was devoted to the event. This time round, the slow and inexorable move towards the same benchmark (and most likely beyond) has been more of a whimper. One reason for that is clearly that there are serious and very obvious risks that markets are keenly focused on. The more important reason is that, similar to the run-up in 2007, oil prices have risen in response to better global growth prospects, first from the improvement in the global growth outlook for emerging economies and more recently in developed ones. The rapid improvement in global growth, in turn, is largely thanks to expansionary monetary and fiscal policies globally. The rise in the oil price today, in other words, is endogenous.
1973: A shock: Treating the oil price rise as an exogenous supply shock akin to the 1973 episode (which is the conventional view of that particular episode - see The Global Monetary Analyst: Seventies Revival, November 17, 2010) would be an incorrect diagnosis and could lead to inappropriate treatment. Back then, central banks in the advanced economies accommodated the shock by quickly cutting interest rates in order to encourage growth. A macro 101 look at this episode using the basic tools of aggregate demand and short-run aggregate supply curves shows the oil shock shifting the supply curve to the left. Central banks then tried to shift the aggregate demand curve to the right by cutting interest rates but were only partially successful. The result was an increase in inflation and inflation expectations without the successful revival of growth. Higher inflation expectations became entrenched and resulted in the stagflation of the 1970s. It took the Fed-engineered recession of 1981 to finally crush inflation expectations and get rid of stagflation.
Today - a symptom: The situation facing EM central banks may appear to be eerily similar to 1973, but is not. Oil prices appear to have risen sharply and some may suggest that EM central banks need to fight this oil ‘shock' by tightening policy faster than is being forecast by analysts and markets. However, if the oil price increase is endogenous, it has been the result of a rightward shift in the aggregate demand curve, thanks to massively expansionary monetary and fiscal policy. In that case, easing the aggregate demand curve back into a position where actual output equals potential should ease the upward pressure on oil prices too. Most EM central banks, particularly ones where output gaps have closed or are rapidly closing, are already moving in that direction.
Bottom line - fundamentals should remain the focus for now: For the different reasons outlined above for food and oil price inflation, EM central banks look set to pay attention primarily to whether their monetary stance is appropriate for the growth profile in place in their economies. Where growth is anaemic and/or core inflation is not at risk, central banks are more likely to look through the supply-driven food inflation and the demand-driven rise in oil prices. On the other hand, signs of rising inflation expectations in China, India and Indonesia, growing attention to inflation risks in Russia and Poland, and domestic demand-led pressure on inflation in Brazil suggest that policy-makers are keenly watching developments, and their policy stance will be set to attenuate or eliminate these risks.
The policy mix - policy rates... Our EM Economics teams expect central banks to continue hiking rates. The rate hikes in 2010 in most places can be broadly considered as a ‘first wave' looking to take policy rates off the historical lows they hit in many countries. Tightening in 2011 is then the ‘second wave' of tightening in response to dwindling slack and the need to accordingly remove some more of the monetary accommodation that is still in place. Only four of the 21 EM central banks (the central banks of Romania, Hong Kong, Malaysia and Mexico) are likely to keep policy rates on hold through 2011. The others are likely to raise policy rates by an average of 90bp over 2011, with the central banks of Chile, Brazil and Turkey leading the way with more than 150bp of rate hikes.
...and exchange rates: Another tool that will help central banks to gain control over inflation is the exchange rate. Currency values in the EM world are generally on an upward trend, thanks to large portfolio flows that have surged into the EM world to take advantage of the two-track nature of the global recovery. EM central banks will thus get a helping hand to deal with increases in imported food and energy prices. While many policy-makers are complaining about the strength of their currencies due to these portfolio flows, a stronger currency will help moderate inflation going forward and portfolio flows could prove to be a blessing in disguise. To the extent that it does the central banks' job by tightening monetary conditions, currency appreciation could allow central banks to hike policy rates at a slower pace (as seems to be the case in Russia and China, for example).
Risks from further shocks to food and oil prices: So far, increases in the price of oil appear to be in line with fundamentals. Our commodity research team expects Brent prices to rise above US$100 in 2011 (see The Commodity Call: Crude Oil: Back to Fundamentals, January 10, 2011). If oil prices rise rapidly and rather suddenly, this will look and feel more like a shock, and central banks would then have to decide how to deal with it. Looking purely at the fundamentals of the policy stance and the growth outlook would no longer be enough at that point. Similarly, if supply shocks to food prices persist, they could start pushing overall inflation expectations higher, raising the risk of higher core inflation.
Dealing with Permanent Shocks
While shocks to food price inflation and oil inflation are likely to be temporary, the shock to the level of food and oil prices may have greater permanence about it. Driven by the structural story of EM outperformance, goods prices that are likely to be in greater demand for the foreseeable future may not necessarily decline in value. This creates risks that monetary policy can and should be vigilant about in the short/medium run, but also advocates quite clearly that there is little that monetary policy can do in the long run.
In the short run, a return of food or oil price inflation to even zero implies that the level of food or oil prices has moved up and stayed up (i.e., we must carefully distinguish disinflation from deflation). This ‘ratcheting up' of the level of prices raises the risk that discretionary income and profits will face sustained headwinds so that households will be more inclined to demand higher nominal wages and firms to pass on costs downstream. Both raise the risk of inflation.
In the long run, however, there is little that central banks can do other than minimise the volatility of inflation around their preferred target (for inflation-targeting central banks). A more certain picture of inflation in the long run should, in theory, lead to better allocation of resources and hence lower volatility of growth.
In summary, central banks are likely to look through transitory, supply-driven shocks to food inflation to the extent that they don't infect core prices. With risks of precisely such a contagion elevated in EM ‘giants' like China, India, Indonesia, Russia, Poland and Brazil, it is safe to say that policy-makers there are likely paying very close attention to inflation developments. In the case of oil prices, their rise so far seems to us to be in line with the recovery in the global economy. In dealing with both food and oil price inflation risks, therefore, EM central banks are likely to pay attention to the slack in the economy and then adjust monetary accommodation accordingly using the exchange rate as well as the policy rate.
The risks to domestic and global inflation are clearly tilted to the upside. Food and oil price inflation may subside but the level of at least some of these prices is likely to remain elevated. If it does, the erosion of disposable income and profit margins might prompt demands for higher nominal wages and higher downstream prices. A large majority of EM central banks will raise policy rates this year, on our forecasts. Rate hikes and appreciating currency values still provide EM central banks with a window of opportunity to overcome inflation. Were Sir Humphrey to resort to vulgar generalisations and crude oversimplifications, his answer to whether central banks are likely to respond aggressively to the recent increases in food and oil prices would be "No, Minister".
Inflation remains >1pp above the BoE's 2.0% target: Much stronger than expected, December CPI inflation rose to 3.7%Y (our and consensus forecast: 3.4%). This reflected upside effects from the air fares component (not expected by us), higher energy prices (as expected) as well as higher food prices (much stronger than we'd penciled in). Given the source, we don't assume any reversal of the upside surprise and expect inflation to rise a bit further in January. We also raise our near-term petrol forecast. All this raises our 2011 inflation forecasts by four-tenths. We still think that CPI inflation will slow sharply in early 2012 as 2011's VAT rise drops out of the year-on-year comparison. But, as our central case, inflation doesn't fall below target at that point. We still think that the near-term balance of risk to our forecasts is to the upside and we remain worried about high inflation in the longer term. The BoE still stands out from other developed economy central banks, with current inflation well above target.
Near-term upside risks: We still see upside risk to our near-term forecasts: inflation expectations are rising; we still see particular upside risks for food and clothing inflation from higher commodity prices; VAT pass-through may well be stronger than we expect.
Implications for the MPC: However, absent a sharp upward wage shock, we do not expect a rate rise until August: 1) Much of the inflation we are seeing is ‘cost-push' inflation and is not good news for the economy. The MPC may see some increased downside risks to its growth forecasts. 2) Hence, the MPC is unlikely to react unless inflation expectations and wages are strongly affected (i.e., prospects for inflation beyond 12 months). We are seeing some signs of the former, but not yet any major pick-up in wage inflation (next average earnings release: January 19). 3) There is likely bad news to come on the activity data as fiscal tightening is reflected (VAT rose only two weeks ago and fiscal spending cuts are set to pick up sharply from April). 4) The December-February data will be even harder to read than usual, given likely substantial weather effects and the VAT rise.
For more detail on our 2011 inflation and monetary policy outlook, see December 2010's edition of The Gilt Edge. We remain concerned about high inflation outcomes longer term and the potential for the 2% target to become untenable (see The Only Way Is Up?... March 31, 2010).
Key Factors and Assumptions Affecting Our Central Inflation Forecasts
Growth: We forecast that GDP growth will be low but positive over 2011, picking up in 2012 (1.6% and 2.0%, respectively). Although spare capacity will dampen inflationary pressures, we think that the amount of spare capacity has already declined significantly.
Interest rates: We expect the BoE to keep interest rates on hold until 3Q11, ending 2011 at 1.0 % and 2012 at 2.0%. Rising interest rates increase the mortgage cost component of RPI and help to keep RPI inflation above CPI inflation over the forecast horizon.
Oil: We have assumed that Brent oil prices rise marginally over the next year and thereafter remain at a high level, which is broadly in line with current futures prices. In sterling terms, using our currency team's GBP/USD forecast, the picture is very similar, but with some small price rises in 2Q12. For details of the currency forecast, see the FX Pulse.
Gas and electricity price rises: We anticipate further increases in gas and electricity prices in 1Q11.
Main Changes over the Past Month
Upside surprise to December CPI inflation is not reversed: We assume that the upside surprise in the food and air fares components is not reversed. This adds two-tenths to our 2011 CPI inflation forecast. We often assume that surprises in air fares will be partly reversed. However, in this case the upside surprise was caused by the increase in weight for this component in 2010 (combined with, as usual, large price rises in December - at nearly 42%M, the same as in December 2009 - ‘exaggerating' the impact of the weight change).
Updated petrol/diesel price profile: Based on incoming data and futures prices, we have revised our petrol/diesel price forecasts. This adds about one-tenth to our 2011 CPI inflation forecast and three-tenths to RPI.
Key Risks
VAT pass-through: January 2011's VAT rise (from 17.5% to 20%) should help to keep inflation above target in 2011. We assume that some VAT pass-through already took place in 4Q10. For January and February 2011, we have assumed around 50% pass-through and a 0.6pp effect on year-on-year inflation (i.e., without the January 2011 VAT hike, inflation would be about 0.6pp lower by February). However, since that is likely similar to the effect we saw from the January 2010 VAT hike, it means that we don't expect year-on-year inflation to increase by much in January. Pass-through could well be stronger than we have penciled in.
Data/methodology changes: January will see CPI weight changes. It will also see a methodology change for items where prices aren't available all year round (e.g., gas BBQs). Currently, the ONS uses the price of the item in the previous ‘observable' month. From January, the ONS will use the current average price movement of the ‘in-season' products. The impact looks likely to range from zero to two-tenths on overall inflation.
Specific risks in clothing inflation: Upside risks stem from higher input costs. Currently, we anticipate average CPI clothing inflation of 0.5%Y in 2011. RPI inflation in clothing and footwear is already 10.3%Y (1.5%Y in CPI).
Higher food price inflation: We continue to worry about the potential for higher food prices in early 2011 after earlier commodity price increases (especially wheat).
Higher inflation expectations and wage demands: A significant upward drift in medium-term household inflation expectations and wage settlements would raise the risk that current high inflation outcomes have serious implications for future inflation. There is more evidence that medium-term household inflation expectations are drifting upwards. However, evidence suggests that pay settlements are only rising modestly and likely higher unemployment will weigh. Nevertheless, there are substantial upside risks: 1) the key period for wage settlements is January to April, when inflation will likely remain well above target; 2) the public sector wage freeze won't weigh on overall earnings growth quite as much as some expect (see UK Labour Market: A Second Softening, November 25, 2010); and 3) consumers will likely have taken a cut in real incomes in 2010 and employees may feel that, after helping their companies work through tough times (e.g., taking unpaid leave), higher wages are justifiable. November average earnings growth, out on January 19, will provide another data point on this. We expect only a one-tenth rise (including bonus, 3M/Y).
Read Full Article »