II. The Globalization of Asian Inflation

International Monetary Fund. Asia and Pacific Dept

Jump to Content Jump to Main Navigation

Author: International Monetary Fund. Asia and Pacific Dept

Language:: English

Keywords:: REO; price; Asia; headline inflation; commodity; core inflation; commodity exporter; inflation in Asia; commodity market imbalance; producer price inflation; Inflation; Commodity prices; Energy subsidies; Consumption; Price controls; Asia and Pacific; Global; consumer price inflation; Oil consumption

Download PDF (530.4 KB)

Abstract

Emerging Asia has recently experienced a sharp rise in inflation, although concerns have now shifted away from inflation onto growth. After reaching double-digit levels in several countries in the region and breaching the target in almost all inflation-targeting regimes, inflation has peaked in most cases through a combination of declining commodity prices, tightened monetary policy, and slowing growth. At the same time, the global financial crisis and concerns that it may translate into a hard landing for global and regional growth have understandably become the main focus of policymakers.

Yet behind this changing short-term focus, Asia has experienced a significant change in the structure of its inflationary process. While in the past, inflation in the region could be traced back to domestic factors and domestic channels, the big increase in global relative prices, notably of commodities, has been the driving force behind rising headline and core inflation this time around. At the same time, increases in commodity prices have been driven in good part by rapid growth in Asia, implying that the region has both exported inflation to the global economy and simultaneously imported it back. In a sense, this has been the counterpart to Asia’s global deflationary impact through the integration of its large and highly productive labor pool into the global economy.

In the near term, sharply falling commodity prices may exert strong deflationary pressures on Asia. At the same time, commodity prices are tentatively expected to return to a high and volatile medium-term equilibrium, the result of underlying imbalances in commodity markets that this chapter will explore. Such gyrations in commodity prices combined with their rising impact on the region’s inflationary process may exacerbate already high inflation volatility, entrench wedges—both positive and negative—between core and headline inflation, and worsen output/inflation volatility trade-offs faced by central banks. Going forward, both the design and conduct of policy in the region will have to take these considerations into account.

Asia’s Changing Inflation Landscape

Inflation in Asia has increased substantially. On a year-on-year basis, emerging Asia’s inflation jumped from 3.6 percent in late 2006 to 7 percent in September 2008, and that in industrial Asia from 0.8 percent to 2.6 percent (Figure 2.1). While some countries, such as Vietnam, Indonesia, and India, have been more affected than others, the increase in inflation has been widespread. Moreover, inflation is now above the target set by the monetary authorities in most inflation-targeting countries. While there are signs that inflation has peaked in some countries, not least thanks to monetary tightening, high producer price inflation coupled with still-biting subsidies suggests that inflationary pressures remain in the system.

More than in the past, inflation in Asia appears to be driven by global factors. Correlations between inflation in Asia and that in the Americas or Europe were already quite high in the early 2000s, at close to 0.5, possibly as a result of monetary policy links (Table 2.1). But in the past four years, inflation in these three regions has been moving in almost perfect tandem, suggesting a common factor underpinning global inflation.

Table 2.1.

Asia Inflation: Correlation with Inflation in Other Regions

(Contemporaneous correlation between estimates of principal components of headline CPI inflation)

Source: IMF staff calculations.

^¹Excludes Argentina and Ecuador.

Table 2.1.

Asia Inflation: Correlation with Inflation in Other Regions

(Contemporaneous correlation between estimates of principal components of headline CPI inflation)

	1999–2002	2003–2008	2007–2008
Americas¹	0.47	0.81	0.92
Europe	0.59	0.91	0.98
World	0.69	0.96	0.98
Commodities	-0.06	0.80	0.97

Source: IMF staff calculations.

^¹Excludes Argentina and Ecuador.

Table 2.1.

Asia Inflation: Correlation with Inflation in Other Regions

(Contemporaneous correlation between estimates of principal components of headline CPI inflation)

	1999–2002	2003–2008	2007–2008
Americas¹	0.47	0.81	0.92
Europe	0.59	0.91	0.98
World	0.69	0.96	0.98
Commodities	-0.06	0.80	0.97

Source: IMF staff calculations.

^¹Excludes Argentina and Ecuador.

Commodity prices, including those for food and fuel, have been the key global factor driving inflation. In recent years, headline inflation in Asia has been very highly correlated with global commodity price inflation. At the domestic level, retail food and fuel prices have contributed between one-third and three-fourths of total cumulative CPI inflation since 2003 (Figure 2.2). This has happened despite below-market prices on these goods in various countries in the region, resulting from the use of direct and indirect subsidies, as well as other administrative measures.

More accurate measurement suggests that imported inflation has accounted for a substantial share of total inflation in Asia, but domestic inflation has also played its part (Figure 2.3). Food and fuel prices need not be a good gauge of true imported inflation, as retail prices for these goods have a large domestic component, while many nonfood, nonfuel items in the CPI can have a large imported component.⁹ Separating for each good and service its underlying imported versus domestic component yields clear patterns:

In commodity-importing countries, which represent most of Asia’s economy, imported inflation has typically contributed about half of total cumulative inflation since 2000. In some economies, like Japan and Taiwan Province of China, estimated imported inflation more than explains inflation, implying that the domestic component may have contributed negatively overall.¹⁰ Moreover, the estimated contribution from imported inflation has risen noticeably over the last two years.¹¹
Nonetheless, the contribution of domestic inflation in commodity-importing economies can be significant, notably in China, where it has accounted for close to 80 percent of total inflation over the last three years.¹²
In commodity-exporting countries such as Australia and New Zealand, the pattern has been different, with domestic inflation estimated to have contributed upwards of 70 percent of cumulative inflation over the period. This is because exchange rate appreciations in some of these countries have dampened the impact of rising import prices, while strong domestic expansions related to terms-of-trade gains have contributed to higher prices in nontradables, such as housing.

It appears inflation in the region has been driven primarily by a large increase in the relative price of commodities, in tandem with inflation elsewhere. With commodity prices falling rapidly in recent months, a key question is whether the run-up that started in 2003 was a temporary or a lasting phenomenon. Part of the answer lies in understanding the role that Asia played in soaring global demand for commodities. This is the focus of our next section.

Asia and the Commodities Boom

Rising demand coupled with inelastic supply contributed to soaring commodity prices in recent years. From 2003 to mid-2008, demand for oil surged under the impetus of emerging markets, while global supply rose only modestly despite the very large price increase. As a result, spare capacity has fallen to historically low levels, at which the impact of small shocks on prices can be disproportionately large, resulting in high price volatility (Figure 2.4). Similar dynamics, albeit less extreme, have characterized other commodity markets.

A key question, then, is whether these demand/supply imbalances are temporary or are here to stay. This question is particularly relevant on account of the sharp drop in commodity prices in recent months. Yet it is worth noting that options markets in early November still signaled close to 25 percent probability that oil prices will be above US$100 at end-2009 despite the global slowdown, and over 10 percent probability that prices will exceed US$120. Most important, options markets signal large uncertainty about the future path of oil prices, implying potentially high volatility (Figure 2.5).

Model estimations also support the view that imbalances have a significant permanent component. These estimations stem from a five-region model of the global economy, which among other features decomposes commodity prices into cyclical output gap, on one side, and permanent changes in the supply/demand balance on the other.¹³ According to these estimates, a substantial share of commodity price inflation in recent years has stemmed from permanent shocks (Figure 2.6), signaling that estimated trend demand has been running faster than estimated trend supply, thereby putting pressures on prices.¹⁴ As others have noted, trend demand from emerging markets is not expected to slow substantially in the medium to long run, while supply has been dogged by lasting factors such as higher extraction costs in marginal fields. Indeed, the “time to build” lags have increased relative to past oil booms, and capacity expansion has consistently fallen short of expectations in recent years.¹⁵ At the same time, cyclical factors played an important role in the latest price run-up before the correction in the third quarter.

What role did Asia play in soaring energy demand? A key one. Given Asia’s rapid energy intensive industrialization, rising incomes, and expanding middle classes, all of which appear to be secular rather than transitory phenomena, it is not surprising (in hindsight) that global energy demand has increased so sharply over the last five years. For instance, global demand for primary energy increased by some 600 million oil equivalent tonnes between 1998 and 2002, but by more than 1,500 million between 2002 and 2007 (Figure 2.7). Asia—mostly China and India—contributed close to two-thirds of the latter increase.

A similar story applies to metals and foods. The cumulative increase in global demand for aluminum and copper—the two key base metals—more than tripled over 2003–2007 relative to 1999–2002 (Figure 2.8). Asia by itself accounted for close to 75 percent of the 2003–2007 increase. Similarly, global demand for soy increased by more than 70 million metric tonnes over 2003–2007, compared with a 50 million metric ton increase in the previous five years. Asia accounted for more than 40 percent of the 70 million metric ton increase. Demand for other grains shows similar patterns.

While rapid trend growth in the global economy contributed to commodity market imbalances, policy distortions also played a part. As is well known, countries in Asia and abroad have kept energy and food prices down through direct and indirect subsidies, preventing demand from fully adjusting to higher international prices. In the case of fuel subsidies, our estimates suggest that removing them throughout the world would have translated into a 2–4 percent drop in global demand for oil (see Box 2.1). Were Asia alone to remove its subsidies, we estimate that global demand would be 1¼–2½ percent lower. While this may not seem large, it is potentially significant in the context of historically low spare capacity. Indeed, a 3 percent lower 2007 global demand would have been equivalent, ceteris paribus, to a doubling of 2007 global spare capacity, restoring some two-thirds of the decline in such capacity since the demand boom started in 2003.

What are the possible policy implications stemming from the underlying changes in Asia’s inflationary process, notably the rising contribution from imported inflation through volatile commodity prices? Addressing this is the purpose of the next section.¹⁶

Policy Considerations

Even in a context where a substantial share of inflation is imported, monetary policy has a key role to play. It is true that shifts in commodity prices have played an important role in fanning inflation in Asia and that these shifts seem not to owe much to global or Asia-specific cyclical considerations. However, monetary policy can ensure that there are no second-round effects from imported inflation on domestic inflation. Beyond these second-round effects, monetary policy has a causal impact on other sources of domestically generated inflation, such as capacity constraints in labor and capital markets. Indeed, the correlation between inflation and model estimates of monetary policy tightness is negative and large in Asia, and negative correlations persist after many quarters, indicating that the effect of monetary policy on inflation is persistent (Figure 2.9).

While monetary policy remains as relevant as ever, changes in Asia’s inflationary process will create new challenges for policy and may entail some reconsideration of policy frameworks. To start, standard practice dictates that monetary policy should not respond to external shifts in relative prices, and contractionary policies should not be implemented to compensate for temporarily higher headline inflation. But standard practice is based on the assumption that the gap between headline and core inflation will be relatively short lived, when in fact the persistence of the commodity price shift has led to large and persistent gaps, in Asia and abroad (Figure 2.10). Going forward, fast declines in commodity prices may now lead to persistent negative gaps between headline and core inflation. Since households and firms ultimately care about headline inflation, persistent gaps, both positive and negative, call into question the usefulness of core inflation as the primary or implicit operational target.

Box 2.1.

Food and Fuel Subsidies in Asia and the World: Are They High, and Have They Affected International Prices?

Subsidies—broadly defined to include administered prices—affect demand and supply through impeding price adjustment. On the supply side, a price ceiling gives little incentive to producers to increase production, while a price floor leads to inefficient overallocation of resources into a sector. From the demand side, direct subsidies and price ceilings lead to overdemand relative to equilibrium. Pervasive subsidization by a group of countries, especially countries in which demand is growing rapidly, could result in market power, pushing global demand, and hence prices, higher.

Direct budget subsidies are a global phenomenon common in both developing and emerging markets, and Asia does not appear to have particularly high direct subsidies by international standards.¹ Nearly one-third of 147 countries sampled by the IMF report direct fuel subsidies in their budgetary accounts, and nearly one-fifth report food subsidies.² The proportion of Asian countries directly subsidizing food—mostly rice—is higher than that in other regions except the Middle East, as is the average fiscal cost, which remains relatively modest. Regarding direct fuel subsidies, the fiscal burden, estimated at 0.1 percent of Asia’s GDP, is small relative to that in other regions. These facts reflect in part that the Asia region is a net oil importer, home to a mere 3 percent of proven reserves. At the same time, the reported Asia average may be artificially low as a result of data limitations in large countries like China. It also masks wide variance within the region, with countries like Indonesia spending more than 2 percent of GDP on fuel subsidies.

Countries with Direct Food and Fuel Subsidies and the Estimated Fiscal Cost, 2007

Sources: Country authorities; and IMF staff estimates.

^¹Share in the regional nominal GDP in U.S. dollars used as weights. Numbers in parentheses specify number of countries covered in a particular region.

Countries with Direct Food and Fuel Subsidies and the Estimated Fiscal Cost, 2007

	Number of Countries Reporting Subsidies in Fiscal Accounts		Weighted Group Average (in percent of regional GDP)¹
	Food subsidies	Fuel subsidies	Food subsidies	Fuel subsidies
Africa (44)	7	8	0.0	0.6
Asia (20)	7	6	0.2	0.1
Europe (26)	0	5	0.0	0.0
Middle East (27)	9	14	0.3	1.4
Americas (30)	4	14	0.0	0.1
Total (147)	27	47	0.1	0.1

Sources: Country authorities; and IMF staff estimates.

^¹Share in the regional nominal GDP in U.S. dollars used as weights. Numbers in parentheses specify number of countries covered in a particular region.

Countries with Direct Food and Fuel Subsidies and the Estimated Fiscal Cost, 2007

	Number of Countries Reporting Subsidies in Fiscal Accounts		Weighted Group Average (in percent of regional GDP)¹
	Food subsidies	Fuel subsidies	Food subsidies	Fuel subsidies
Africa (44)	7	8	0.0	0.6
Asia (20)	7	6	0.2	0.1
Europe (26)	0	5	0.0	0.0
Middle East (27)	9	14	0.3	1.4
Americas (30)	4	14	0.0	0.1
Total (147)	27	47	0.1	0.1

Sources: Country authorities; and IMF staff estimates.

^¹Share in the regional nominal GDP in U.S. dollars used as weights. Numbers in parentheses specify number of countries covered in a particular region.

An alternative and in some sense broader measure of fuel subsidies compares the domestic retail fuel price to the international price, with a higher price implying a net tax and a lower price indicating a net subsidy.³ About 17 percent of the 127 countries for which retail gasoline prices are available for end-2007 have a retail price below the world price, with many of these countries in the Middle East; a somewhat higher percentage of countries have net subsidies on diesel. Calculating subsidies on this basis, it does not appear that Asia has higher subsidies than the rest of the world—indeed the average net tax in Asia (including countries with net subsidies) is higher than that in the United States. Net taxes in fast-growing China, while positive, are significantly below the world average.

Fuel Subsidy Estimates Based on Retail Price Data, End-2007

Sources: IMF, Fiscal Affairs Department, based on desk submissions, and staff calculations.

^¹The world price reflects complete pass-through plus processing and distribution costs, but net of any taxes or subsidies.

Fuel Subsidy Estimates Based on Retail Price Data, End-2007

	Number of Countries for Which Retail Data Are Available		Number of Countries with Retail Price Less Than World Price¹		Percent Subsidy (average)		Number of Countries with Retail Price More Than World Price¹		Percent Tax (average)
	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel
Africa	37	35	4	4	-23.3	-39.5	33	31	135.0	91.7
Asia	16	15	2	3	-10.7	-26.8	14	12	85.9	43.4
Europe	32	32	0	0	0.0	0.0	32	32	201.3	145.6
Middle East	26	28	14	21	-41.9	-56.8	12	7	39.9	28.1
Americas	16	15	2	4	-12.5	-29.8	14	11	66.1	38.9
Total	127	125	22	32	-22.1	-38.2	105	93	105.6	69.6

Sources: IMF, Fiscal Affairs Department, based on desk submissions, and staff calculations.

^¹The world price reflects complete pass-through plus processing and distribution costs, but net of any taxes or subsidies.

Fuel Subsidy Estimates Based on Retail Price Data, End-2007

	Number of Countries for Which Retail Data Are Available		Number of Countries with Retail Price Less Than World Price¹		Percent Subsidy (average)		Number of Countries with Retail Price More Than World Price¹		Percent Tax (average)
	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel	Gasoline	Diesel
Africa	37	35	4	4	-23.3	-39.5	33	31	135.0	91.7
Asia	16	15	2	3	-10.7	-26.8	14	12	85.9	43.4
Europe	32	32	0	0	0.0	0.0	32	32	201.3	145.6
Middle East	26	28	14	21	-41.9	-56.8	12	7	39.9	28.1
Americas	16	15	2	4	-12.5	-29.8	14	11	66.1	38.9
Total	127	125	22	32	-22.1	-38.2	105	93	105.6	69.6

Sources: IMF, Fiscal Affairs Department, based on desk submissions, and staff calculations.

^¹The world price reflects complete pass-through plus processing and distribution costs, but net of any taxes or subsidies.

Data on indirect subsidies through the use of administered prices for fuels sold to industry are very limited, but anecdotal evidence suggests that such subsidies also span the globe. In addition to Asia, indirect subsidies can be found in Latin America, the Middle East oil-producing nations, and Eastern Europe/Central Asia. Indirect subsidies usually imply direct or close participation by the government in the production and/or distribution process. In some cases, governments directly control import levels, domestic distribution, and domestic prices. In others, government-owned or government-linked companies sell their products at a ceiling price without direct budgetary compensation, depressing profit margins and leading to quasi-fiscal losses to revenue.

There is some evidence that fuel subsidies, direct and indirect, have contributed to higher demand for oil. The rate of growth of consumption of both primary fuels and oil between 2003 and 2007 in a group of subsidizing countries chosen for study⁴ was five times that of non-subsidizing countries (18.9 percent versus 3.6 percent), in part driven by very high demand growth in China. As a result, subsidizing countries’ share of primary fuels consumption has increased by nearly 5 percentage points since 2003, and by about half that amount for oil. Even taking into consideration low GDP and income growth in non-subsidizing OECD countries versus subsidizing emerging markets, the difference in demand growth is sizeable, pointing to a likely impact of subsidy policies on consumption.

Moreover, fuel subsidies may have had enough of an impact on global demand to affect market balances materially. To start, it is worth noting that OECD countries, which typically do not subsidize fuels, still account for more than 50 percent of total primary fuels consumption and just under 60 percent of global oil consumption. Moreover, only about half of the non-OECD fuel consumption is taken up by emerging markets reported to have sizeable subsidies. This being said:

If oil consumption in subsidizing countries had grown at the same rate as that in non-subsidizing emerging countries, global demand in 2007 would have been 1½ percent lower. If demand from subsidizing countries had grown at the same rate as that in all non-subsidizing countries, emerging and developed, global demand would have been 3½ percent lower.
If it is assumed that the domestic price in subsidizing countries is, on average, one-third to one-half of the full pass-through price, and that the short-run elasticity of oil demand is between 5 percent and 10 percent, total global consumption without subsidies would have been 1½–5½ percent lower than the 2007 outturn.⁵
While demand in subsidizing countries rose rapidly, net exports by these countries also rose if one excludes China, with increased production by Russia, and to a lesser extent Brazil, Ecuador, and Iran, more than offsetting increases in demand and in some cases declines in production elsewhere. When China is included, net exports of subsidizing countries declined by 67 million tonnes, less than 2 percent of global 2007 consumption.

Growth in Consumption of Oil and Primary Fuels, 2003–07

(In percent)

Sources: Statistical Review of World Energy, 2008, BP PLC.; and IMF staff calculations.

Growth in Consumption of Oil and Primary Fuels, 2003–07

(In percent)

		Oil			Primary Fuels
		2003–07	2003	2007	2003–07	2003	2007
		Change	Share	Share	Change	Share	Share
World		7.4	100.0	100.0	12.9	100.0	100.0
Selected subsidizing countries		18.9	24.7	27.3	30.6	31.0	35.8
	excl. China	12.2	14.9	15.6	16.4	16.6	17.1
World excluding selected subsidizers		3.6	75.3	72.7	5.0	69.0	64.2
OECD countries		1.2	60.4	56.9	2.7	55.2	50.2

Sources: Statistical Review of World Energy, 2008, BP PLC.; and IMF staff calculations.

Growth in Consumption of Oil and Primary Fuels, 2003–07

(In percent)

		Oil			Primary Fuels
		2003–07	2003	2007	2003–07	2003	2007
		Change	Share	Share	Change	Share	Share
World		7.4	100.0	100.0	12.9	100.0	100.0
Selected subsidizing countries		18.9	24.7	27.3	30.6	31.0	35.8
	excl. China	12.2	14.9	15.6	16.4	16.6	17.1
World excluding selected subsidizers		3.6	75.3	72.7	5.0	69.0	64.2
OECD countries		1.2	60.4	56.9	2.7	55.2	50.2

Sources: Statistical Review of World Energy, 2008, BP PLC.; and IMF staff calculations.

These three approaches suggest that, in the absence of subsidies across all countries, global 2007 demand would have been some 2–4 percent lower. Were Asia alone to remove its subsidies, global 2007 demand would have been some 1¼–2½ percent lower, given that Asia has accounted for a large share in global demand growth.⁶ While these numbers may not seem large, it is worth noting that global oil markets have in recent years been characterized by historically low spare capacity, which in turn may have been a cause of high oil prices or a symptom of prospective oil scarcity. Indeed, a 3 percent lower 2007 global demand would have been equivalent, ceteris paribus, to a doubling of 2007 global spare capacity, restoring some two-thirds of the decline in such capacity since the demand boom started in 2003.

¹ Direct subsidies are found most often in countries with liberalized product import and/or distribution, with the government setting a ceiling price then compensating private sector firms for the price differential plus an agreed margin.² IMF, “Food and Fuel Prices—Recent Developments, Macroeconomic Impact, and Policy Responses,” available via the Internet: www.imf.org/external/np/exr/foodfuel/index.htm.³ The world price reflects complete pass-through plus processing and distribution costs, but net of taxes and subsidies.⁴ Information on indirect subsidies is elusive, hence the need to focus on a selected group of countries. The selected 13 emerging countries are Argentina, Brazil, China, Ecuador, Egypt, India, Indonesia, Iran, Malaysia, Pakistan, Russia, Saudi Arabia, and Venezuela. They were chosen among all subsidizing countries because of their relatively large contribution to global energy demand.⁵ Recent studies find that the short-run elasticity of oil demand has fallen dramatically over the past few decades and is now very low, in the range of less than 10 percent. See Hamilton (2008) and Cooper (2003).⁶ This calculation is derived using the simplifying assumption that subsidies in Asia had the same impact on Asian demand as subsidies in other regions had on demand in those regions.

With inflation increasingly driven by external shocks, policymakers may face a more difficult trade-off between output and inflation stabilization. In normal times, most central banks in the region pursue the dual objectives of output and inflation stabilization around a certain target.¹⁷ When inflation is driven primarily by shocks to domestic demand, the two objectives may be fully compatible, as stimulating/slowing economic growth would a priori lead to higher/lower inflation. But when external price shifts—both positive and negative—play an important role in the inflationary process, policymakers may be forced to choose between stabilizing inflation and stabilizing output around their targets, in particular in commodity-importing countries. For instance, exogenous increases in commodity prices could feed into headline inflation at a time when a country’s output is below potential. To quantify these issues, we looked at what would happen to the volatility of inflation around its target in the context of larger (in absolute value) commodity shocks, assuming central banks in the region keep constant the volatility of output around potential. In almost all countries in the region, inflation volatility would go up (Figure 2.11); conversely, were central banks to keep inflation volatility constant under stronger commodity price shocks, output volatility would go up.¹⁸

Moreover, highly volatile commodity prices are likely to translate into more volatile inflation. Indeed, there is evidence that inflation volatility has gone up in Asia in recent years, with the standard deviation increasing by around 30 percent since 2005 (Table 2.2). While inflation volatility may have increased as a result of other factors, it is noteworthy that commodity-importing countries have experienced the highest jump in volatility. The more volatile inflation is, the harder it is for economic agents to predict it correctly, leading to greater resource misallocation.

Table 2.2.

Volatility of Inflation

Sources: IMF, Information Notice System, World Economic Outlook, and staff calculations.

^¹Commodity exporters are Australia, New Zealand, Indonesia, and Malaysia.

^²Commodity importers are Japan, Hong Kong SAR, Korea, Singapore, Taiwan POC, China, India, the Philippines, Thailand, and Vietnam.

Table 2.2.

Volatility of Inflation

Economy		Standard Deviation
Economy		2000–2004	2005–July 2008
Asia		1.57	2.06
	Industrial Asia	0.78	0.69
	Emerging Asia	1.78	2.43
	Commodity exporters¹	1.55	1.77
	Commodity importers²	1.58	2.18

Sources: IMF, Information Notice System, World Economic Outlook, and staff calculations.

^¹Commodity exporters are Australia, New Zealand, Indonesia, and Malaysia.

^²Commodity importers are Japan, Hong Kong SAR, Korea, Singapore, Taiwan POC, China, India, the Philippines, Thailand, and Vietnam.

Table 2.2.

Volatility of Inflation

Economy		Standard Deviation
Economy		2000–2004	2005–July 2008
Asia		1.57	2.06
	Industrial Asia	0.78	0.69
	Emerging Asia	1.78	2.43
	Commodity exporters¹	1.55	1.77
	Commodity importers²	1.58	2.18

Sources: IMF, Information Notice System, World Economic Outlook, and staff calculations.

^¹Commodity exporters are Australia, New Zealand, Indonesia, and Malaysia.

^²Commodity importers are Japan, Hong Kong SAR, Korea, Singapore, Taiwan POC, China, India, the Philippines, Thailand, and Vietnam.

These challenges will put a premium on good central bank communication. With highly volatile inflation, clear communication by central banks will be key to help anchor expectations. Central banks, and especially inflation targeters, need to communicate the time frame within which they expect shocks (positive and negative) to inflation to subside and inflation to return to its implicit or explicit target; what assumptions underpin this view; and what policy measures would be undertaken in the event this baseline scenario does not materialize. This is particularly true in countries where monetary policy credibility still remains to be fully established.

Should the endogeneity of commodity prices be taken into account by policymakers? As discussed, high commodity prices have resulted in part from strong underlying growth in Asia, begging the important question of whether commodity prices should be part of the policy reaction function of central banks. In countries whose effect on global commodity prices is negligible, it seems uncontroversial that they should not be. But even in the larger countries, “targeting” of commodity prices would appear misplaced given the global nature of commodity markets and the fact that imbalances in these markets seem more secular than policy driven. At the same time, commodity prices seem to contain useful information for monetary policy. In particular, higher commodity-driven inflation in recent years at a time when wage data and other indicators signaled little in the way of capacity pressures seems to suggest that energy prices are a better indicator of the output gap, in Asia and globally. Conversely, falling commodity prices appear to be a leading indicator of future capacity constraints, ahead of labor market data.

Concluding Remarks

Beyond short-term considerations in regard to the expected path of inflation in various Asian countries, the nature of inflation in Asia has changed substantially over the last several years. In particular, global relative price shifts, both positive and negative, are now playing a larger role. Cyclically low commodity prices are tentatively expected to return to a high and volatile medium-term equilibrium, the result of underlying imbalances in commodity markets. These trends have important implications for the design and conduct of monetary policy in the region, as well as the challenges and trade-offs that policymakers are likely to face.

Appendix

Decomposing Imported and Domestic Components of Inflation

We perform this decomposition not by arbitrarily classifying each goods category in the CPI basket as “imported” or “domestic,” but by proxying the CPI index through a convex combination of the import price deflator and the GDP deflator, with weights equal to the import content of consumption and one minus the import content of consumption, respectively. Such a decomposition is at the root of most modern open macro models, as can be seen in Obstfeld and Rogoff (1996). When not available, the respective weights were estimated through regression analysis to minimize the distance between the proxy and the actual CPI. In Asian countries, the convex combinations track the actual CPI quite closely.

Estimated Model of the Global Economy

Our econometric analysis is based on an estimated unobserved-components model of the world economy featuring five open economies connected by trade and financial linkages. Cyclical components are modeled as a multivariate linear rational expectations model of the monetary transmission mechanisms in these five economies. The monetary transmission mechanism in each economy is described by an aggregate supply relationship, an aggregate demand relationship, a nominal interest rate rule, and a generalized real uncovered interest parity condition. Trend components are modeled as independent random walks.

In what follows, $x_{i, t}^{f}$ denotes the arithmetic trade-weighted average of variable x_i,t across the trading partners of economy i, while $x_{t}^{w}$ denotes the arithmetic output-weighted average of variable x_i,t across all economies. Furthermore, E_t x_{i, t + s} denotes the cyclical component of variable x_{i, t + s} associated with economy i, conditional on information available at time t. Finally, ${\hat{x}}_{i, t}$ denotes the cyclical component of variable x_i,t while ${\bar{x}}_{i, t}$ denotes the trend component of variable x_i,t. Cyclical and trend components are additively separable, that is, $x_{i, t} = {\hat{x}}_{i, t} + {\bar{x}}_{i, t}$ .

Cyclical Components

The cyclical component of consumption price inflation ${\hat{π}}_{i, t}$ depends on a linear combination of its past and expected future cyclical components driven by the contemporaneous cyclical component of output according to the aggregate supply relationship,

\begin{matrix} \begin{array}{l} {\hat{π}}_{i, t} = ø_{1, 1} {\hat{π}}_{i, t - 1} + ø_{1, 2} E_{t} {\hat{π}}_{i, t + 1} θ_{1, 1} \ln {\hat{Y}}_{i, t} \\ + θ_{1, 2} \frac{M_{i}}{Y_{i}} [Δ \ln {\hat{Q}}_{i, t} - ø_{1, 1} Δ \ln {\hat{Q}}_{i, t - 1} - ø_{1, 2} E_{t} Δ \ln {\hat{Q}}_{i, t + 1}] \\ + θ_{1, 3} \frac{M_{i}}{Y_{i}} [Δ \ln ({\hat{S}}_{i, t}^{U S A} {\hat{P}}_{t}^{C O M}) - ø_{1, 1} Δ \ln ({\hat{S}}_{i, t - 1}^{U S A} {\hat{P}}_{t - 1}^{C O M}) - ø_{1, 2} E_{t} Δ \ln ({\hat{S}}_{i, t + 1}^{U S A} {\hat{P}}_{t + 1}^{C O M})] + v_{i, t}^{\hat{P}}, \end{array} & (2 A .1) \end{matrix}

where supply shock $v_{i, t}^{\hat{P}} = ρ_{\hat{P}} v_{i, t - 1}^{\hat{P}} + ε_{i, t}^{\hat{P}}$ with $ε_{i, t}^{\hat{P}} ~ iid N (0, σ_{\hat{P}, i}^{2})$ . The cyclical component of consumption price inflation also depends on contemporaneous, past, and expected future changes in the cyclical components of the real effective exchange rate and the domestic-currency-denominated price of commodities. The response coefficients of this relationship vary across economies with their degree of openness, measured by the ratio of imports to output $\frac{M_{i}}{Y_{i}}$ .

The cyclical component of output In ${\hat{Y}}_{i, t}$ depends on a linear combination of its past and expected future cyclical components, driven by the contemporaneous cyclical component of the real interest rate according to the aggregate demand relationship,

\begin{matrix} \begin{array}{l} \ln {\hat{Y}}_{t, t} = ø_{2, 1} {\hat{Y}}_{t, t - 1} + ø_{2, 2} E_{t} \ln {\hat{Y}}_{t, t + 1} θ_{2, 1} (1 + \frac{M_{t}}{Y_{t}})^{- 1} {\hat{r}}_{t, t} \\ + θ_{2, 2} \frac{X_{i}}{Y_{i}} (1 + \frac{M_{t}}{Y_{t}})^{- 1} [\ln {\hat{Y}}_{t, t}^{f} - ø_{2, 1} {\hat{Y}}_{t, t - 1}^{f} - ø_{2, 2} E_{t} Δ \ln {\hat{Y}}_{t, t + 1}^{f}] \\ + θ_{2, 3} \frac{X_{t} + M_{i}}{Y_{i}} (1 + \frac{M_{t}}{Y_{t}})^{- 1} [\ln {\hat{Q}}_{t, t} - ø_{2, 1} \ln {\hat{Q}}_{t, t - 1} - ø_{2, 2} E_{t} \ln {\hat{Q}}_{t, t + 1}] + v_{t, t}^{\hat{Y}}, \end{array} & (2 A .2) \end{matrix}

where demand shock $v_{i, t}^{\hat{Y}} = ρ_{\hat{Y}} v_{i, t - 1}^{\hat{Y}} + ε_{i, t}^{\hat{Y}}$ with $ε_{i, t}^{\hat{Y}} ~ iid N (0, σ_{\hat{Y}, i}^{2})$ . Reflecting the existence of both interest rate and exchange rate channels of monetary policy, the cyclical component of output also depends on the contemporaneous, past, and expected future cyclical components of foreign output and the real effective exchange rate. The response coefficients of this relationship vary across economies with their degree of openness, measured by the ratio of exports to output $\frac{X_{i}}{Y_{i}}$ or imports to output $\frac{M_{i}}{Y_{i}}$ .

The cyclical component of the nominal interest rate ${\hat{i}}_{i, t}$ depends on a weighted average of its past and desired cyclical components according to the monetary policy rule,

\begin{matrix} \begin{array}{l} {\hat{i}}_{i, t} = ø_{3, 1} {\hat{i}}_{i, t - 1} \\ + (1 - ø_{3 - 1, i}) (θ_{3, 1, i} {\hat{π}}_{i, t} + θ_{3, 2, i} \ln {\hat{Y}}_{i, t} θ_{3, 3, i} {\hat{i}}_{U S A, t} + θ_{3, 4, i} \ln {\hat{S}}_{i, t}^{U S A}) + ε_{i, t}^{\hat{i}}, \end{array} & (2 A .3) \end{matrix}

where monetary policy shock $ε_{i, t}^{\hat{i}} ~ N (0, σ_{\hat{i}, t}^{2})$ . The desired cyclical component of the nominal interest rate depends on the contemporaneous cyclical components of consumption price inflation, output, a foreign nominal interest rate, and the nominal bilateral exchange rate. The response coefficients of this rule vary across economies, nesting flexible inflation targeting and fixed exchange rate regimes as special cases of the conduct of monetary policy. The cyclical component of the real interest rate ${\hat{r}}_{i, t}$ satisfies ${\hat{r}}_{i, t} = {\hat{i}}_{i, t} - E_{t} {\hat{π}}_{i, t + 1}$ .

The cyclical component of the real bilateral exchange rate In ${\hat{Q}}_{i, t}^{U S A}$ depends on a linear combination of its past and expected future cyclical components driven by the contemporaneous cyclical component of the real bilateral interest rate differential,

\begin{matrix} \ln {\hat{Q}}_{i, t}^{U S A} = ø_{4, 1} \ln {\hat{Q}}_{i, t - 1}^{U S A} + ø_{4, 2} E_{t} \ln {\hat{Q}}_{i, t + 1}^{U S A} + θ_{4, 1} ({\hat{r}}_{i, t} - {\hat{r}}_{U S A, t}) + {\hat{ε}}_{i, t}^{\hat{S}}, & (2 A .4) \end{matrix}

where risk premium shock ${\hat{ε}}_{i, t}^{\hat{S}} ~ iid N (0, σ_{\hat{S}, t}^{2})$ . The cyclical component of the real bilateral exchange rate also satisfies ${\hat{Q}}_{i, t}^{U S A} = \ln {\hat{S}}_{i, t}^{U S A} + \ln {\hat{P}}_{i, t}^{U S A} - \ln {\hat{P}}_{i, t}$ where In ${\hat{S}}_{i, t}^{U S A}$ denotes the cyclical component of the nominal bilateral exchange rate.

The cyclical component of the change in the price of commodities ${\hat{P}}_{t}^{C O M}$ depends on a linear combination of its past and expected future cyclical components driven by the contemporaneous cyclical component of world output,

\begin{matrix} Δ \ln {\hat{P}}_{t}^{C O M} = ø_{5, 1} Δ \ln {\hat{P}}_{t - 1}^{C O M} + ø_{5, 2} Δ \ln {\hat{P}}_{t + 1}^{C O M} + θ_{5, 1} Δ \ln {\hat{Y}}_{t}^{w} + {\hat{ε}}_{t}^{{\hat{P}}^{C O M}}, & (2 A .5) \end{matrix}

where commodity price shock $ϵ_{t}^{{\hat{P}}^{C O M}} ~ iid N (0, σ_{{\hat{P}}^{C O M}}^{2})$ . As an identifying restriction, all innovations are assumed to be independent.

Trend Components

The growth rates of the trend components of the price of consumption, In ${\bar{P}}_{i, t}$ , output In ${\bar{Y}}_{i, t}$ , and the price of commodities In ${\bar{P}}_{t}^{C O M}$ follow random walks:

\begin{matrix} Δ \ln {\bar{P}}_{i, t} = Δ \ln {\bar{P}}_{i, t - 1} + ε_{i, t}^{\bar{P}}, ε_{i, t - 1}^{\bar{P}} ~ iid N (0, σ_{\bar{P}, i}^{2}), & (2 A .6) \end{matrix}

\begin{matrix} Δ \ln {\bar{Y}}_{i, t} = Δ \ln {\bar{Y}}_{i, t - 1} + ε_{i, t}^{\bar{Y}}, ε_{i, t - 1}^{\bar{Y}} ~ iid N (0, σ_{\bar{Y}, i}^{2}), & (2 A .7) \end{matrix}

\begin{matrix} Δ \ln {\bar{P}}_{t}^{C O M} = Δ \ln {\bar{P}}_{t - 1}^{C O M} + ε_{i, t}^{{\bar{P}}^{C O M}}, ε_{i, t - 1}^{{\bar{P}}^{C O M}} ~ iid N (0, σ_{{\bar{P}}^{C O M}}^{2}) . & (2 A .8) \end{matrix}

The trend components of the nominal interest rate ${\bar{i}}_{i, t}$ and nominal bilateral exchange rate $\ln {\bar{S}}_{i, t}^{U S A}$ also follow random walks:

\begin{matrix} {\bar{i}}_{i, t} = {\bar{i}}_{i, t - 1} + ε_{i, t}^{\bar{i}}, ε_{i, t}^{\bar{i}} ~ iid N (0, σ_{\bar{i}, i}^{2}), & (2 A .9) \end{matrix}

\begin{matrix} \ln {\bar{S}}_{i, t}^{U S A} = \ln {\bar{S}}_{i, t - 1}^{U S A} + ε_{i, t}^{\bar{S}}, ε_{i, t}^{\bar{S}} ~ iid N (0, σ_{\bar{S}, i}^{2}) . & (2 A .10) \end{matrix}

The trend component of the real interest rate ${\bar{r}}_{i, t}$ satisfies ${\bar{r}}_{i, t} = {\bar{i}}_{i, t} - E_{t} {\bar{π}}_{i, t + 1}$ while the trend component of the real bilateral exchange rate In ${\bar{Q}}_{i, t}^{U S A}$ satisfies $\ln {\bar{Q}}_{i, t}^{U S A} = \ln {\bar{S}}_{i, t}^{U S A} + \ln {\bar{P}}_{U S A, t} - \ln {\bar{P}}_{t, t}$ . As an identifying restriction, all innovations are assumed to be independent.

Estimation Procedure

The parameters and unobserved components of this unobserved-components model are jointly estimated with a Bayesian procedure, conditional on prior information concerning the values of parameters and trend components. Inference on the parameters is based on an asymptotic normal approximation to the posterior distribution around its mode, which is calculated by numerically maximizing the posterior density kernel.

Evaluation of the posterior density kernel involves first constructing a multivariate linear rational expectations representation of those equations governing the evolution of cyclical components, then solving for the unique stationary solution to this multivariate linear rational expectations model with the algorithm due to Klein (2000). The resultant first-order vector autoregressive representation of those equations governing the evolution of cyclical components is then combined with a dynamic factor representation of those equations governing the evolution of trend components to form a linear state space model. This linear state space model is then augmented with a set of stochastic restrictions summarizing prior information concerning the values of the trend components. The predictive density function is then evaluated, conditional on the parameters associated with this linear state space model, with an adaptation of the filter due to Kalman (1960) that incorporates prior information. Finally, this conditional density function is combined with a multivariate normal density function summarizing prior information concerning the values of parameters.

Description of the Data Set

The data set consists of quarterly observations on several macroeconomic and financial market variables for seven economies over the period 1999:Q3 through 2008:Q1. The economies under consideration are China, the European Union, India, Japan, Korea, Thailand, and the United States.

The macroeconomic variables under consideration are the price of consumption, output, and the price of commodities. The price of consumption is proxied by the seasonally adjusted total consumer price index. Output is measured by seasonally adjusted real gross domestic product. The price of commodities is proxied by a broad commodity price index denominated in U.S. dollars. All macroeconomic data were obtained from the IMF’s World Economic Outlook database.

The financial market variables under consideration are the nominal interest rate and the nominal bilateral exchange rate. The nominal interest rate is measured by the money market rate expressed as a period average where possible, and the bank rate quoted as an end-of-period value where necessary. The nominal bilateral exchange rate is measured by the domestic currency price of one U.S. dollar quoted as an end-of-period value. All financial market data were extracted from the IMF’s International Financial Statistics database.

Topics

Countries

Series

About

Resources

Abstract

Asia’s Changing Inflation Landscape

Asia and the Commodities Boom

Policy Considerations

Concluding Remarks

Appendix

Decomposing Imported and Domestic Components of Inflation

Estimated Model of the Global Economy

Cyclical Components

Trend Components

Estimation Procedure

Description of the Data Set

Within Same Series

Other IMF Content

Other Publishers Content

Asian Development Bank

Food and Agriculture Organization

Inter-American Development Bank

The World Bank

Share Link

Table of Contents

Sections

Figures

References

Metrics