Inflation and Conflict in Iraq: The Economics of Shortages Revisited

Starting in December 2005, with further adjustments in the second half of 2006 and in 2007, the price of regular gasoline has been increased from less than 3 U.S. dollar cents per liter to 32 U.S. dollar cents per liter, above the average in other oil-exporting countries in the region. Similar price adjustments have also been implemented for premium gasoline, diesel, LPG, and kerosene.

The average ratio of black market to administered prices declined from approximately 4.3 following the December 2005 adjustment to 2.8 following the January 2007 adjustment.

Inflation refers to both off-equilibrium price increases (captured in the model by shortages) as well as increases in the equilibrium price level (captured in the model by violence, a proxy for expectations of future policy changes).

Historically, Iraq is not a high-inflation country. Foote, Block, and Gray (2004) point out that between 1945 and 1989 inflation usually stayed in the single digits. Only after the first Gulf war and the following UN Embargo did inflation surge to nearly 500%, as the government began to monetize its fiscal deficits.

The Iraqi Consumer Price Index includes both black market and official prices for gasoline, calculated using 74 and 26% weights, respectively, reflecting the estimated shares of fuel consumption out of black and official markets in total.

By early 2009 fuel shortages had been reduced significantly with only 32% of Iraqi’s reporting fuel shortages in their neighborhood, compared to 81% in March 2008 (Source: BCC/ABC Iraq Poll, http://news.bbc.co.uk/2/shared/bsp/hi/pdfs/13_03_09_iraqpollfeb2009.pdf).

The absence of noticeable pressures on the exchange rate in the period after the introduction of the new Iraqi Dinar (in 2003) is noteworthy. As King (2004) points out, “...the value of money depends on beliefs about the probability of survival of the institutions that define the state itself”. Apparently, for most Iraqi’s the survival of the post-Sadam state was never in doubt, despite the rising violence.

Some authors see the payout of large subsidies as a disciplinary device for public finances (Rowat 2008). In countries where the government derives most of its revenues from rents (mostly from natural resources, as in the case of Iraq) instead of tax revenues, subsidies for virtually the entire population serve as a legitimate national dividend payout.

A law was enacted in September 2006 to allow the private sector to import fuel. However, imports of private sector gasoline did not begin until late-2007.

Iraq Poll 2007, BBC News.

“Even Green Zone Has It Own Fuel Crisis,” Azzaman newspaper, July 4, 2007.

Although more difficult to measure, other goods have also been in short supply as a result of the violence. According to the BBC survey, 62% of Iraqi respondents reported that availability of basic goods needed for their households was “bad”, up from 38% in 2005 when the poll was last conducted.

Because all operations of the oil sector are linked, a disruption in one part can shut down a different part of the oil infrastructure. For example, if the flow of fuel oil, a by-product from the refinery process, from the storage tanks to the power plants or the export terminals is interrupted, this not only interrupts electricity supply or fuel exports but also disrupts refinery operations and the domestic supply of gasoline because of capacity constraints of storage tanks for fuel oil.

The assumption of a Leontieff-type technology for production (and distribution) of non-oil goods is plausible here given that the technology is fixed in the short run and is unlikely to change drastically given the security situation.

The marginal revenue curve is based on eq. [2]. We assume a second-order quadratic specification to fit the data better.

Due to lack of suitable proxies with monthly frequency we did not include any measures of economic activity in the regressions. (Data comprising Iraq’s GDP are poorly measured and are only available on an annual basis). This omission is likely to have only minor implications, given the short period under review.

Note that the administered prices have a relatively small weight in the calculation of the fuel price index in Iraq (26%, with the remainder being accounted for by black market prices), which itself has only a 2.1% weight in the overall basket, and are unlikely to serve as true price/reference point for fuel products for the vast majority of market participants. This should limit the risk of direct collinearity between administered prices and inflation measures.

Chaney (2008) uses the same indicators as control variables in an analysis that uses the variation in the Iraqi sovereign bond spreads to evaluate pacification policy in Iraq.

To the extent that the exchange rate has been kept fixed for most of the period under review (and the capital account was free), it is possible that the money supply was endogenously determined by nominal income/prices, rather than the other way around. Absent reliable and long time series on income measures, we could not test this in the context of this study. We do, however, expect that lagging the reserve money should reduce the extent of this endogeneity problem. Related to this, although the pervasive dollarization would make the growth of dinar money supply an imperfect predictor of inflation (as it would understate the extent of total – that is, dinar plus dollar – money supply growth), it was practically impossible for us to find an adequate measure of cash dollars circulating in the economy. Hence we used only dinar reserve money as a measure of money supply.

A point regarding the quality of inflation data in Iraq is warranted at this juncture. It is quite conceivable that divergence of violence across regions had implications on the ability of staff in regional statistical offices to secure accurate data. Therefore, while to the best of our knowledge the data we use have consistent geographical coverage, to the extent that the violence may have had implications on the quality of data collected, the country-wide inflation data might somewhat inaccurately describe price movements across the country. However, at this point we are unable to either test or control for this (potentially minor) effect in our analysis.

Because of short time span of our observation period, all possible adjustments in the short-run behavior of prices due to long-term misalignment in either the money or foreign exchange markets are very difficult to measure and have been assumed negligible.

Measuring the domestic component of the fiscal envelope proved impossible.

The signs of the quarterly dummies in the equations of core inflation indicate that inflation is higher in Q1, by about 2%, which could reflect a longer pass-through from supply to price adjustments for food and other non-fuel products than for fuel products.