Measuring inflation more effectively

For Tetum, full texts, and additional information, click here.

Timor-Leste's National Statistics Directorate (DNE) publishes Consumer Price Index (CPI) reports to measure inflation in Timor-Leste. They are updating their 11-year-old system with technical help from the Australian Bureau of Statistics. Using information from the (as-yet-unpublished) 2011 Household Income and Expenditure Survey (HIES), they will change the weights for different product groups and also hope to produce a monthly national CPI. DNE recently organized a workshop where two ABS experts explained how CPI is derived and presented proposed changes. Following a lively discussion, DNE encouraged participants to share additional thoughts, so La’o Hamutuk wrote the following:

La'o Hamutuk Submission on revising TL's Consumer Price Index (abridged)

21 November 2012

Download PDF of La'o Hamutuk's complete submission in English or Tetum.

Thank you for encouraging La’o Hamutuk to give suggestions. We believe that a solid understanding of today is essential to developing good strategies for tomorrow, and are glad that DNE has asked users of your publications for input.

We are not experts in statistics, although we see CPI as an important tool for advancing economic justice. Measuring inflation is a challenging undertaking which cannot be reduced to a single monthly number, and we hope our observations will help policy makers understand and control the increasing cost of living for Timor-Leste’s citizens.

Timor-Leste’s economy and consumption patterns are very different from Australia. Timor-Leste has a diverse  economy, and our small size makes variations more extreme. In addition to the obvious differences for the 25% of Timorese who live in Dili, there are differences among and within the districts. Those which border Indonesia purchase more “informal” imports, while those which produce coffee have seasonal variations. In each district, people who live in towns buy differently from those in more remote areas. Therefore, we worry that increasing the frequency of national CPI statistics while reducing the number of places sampled could introduce inaccuracies.

We appreciate that the CPI measures effects on households. In an economy like Timor-Leste, where a small fraction of people has most of the income, indicators based on dollars (such as GDP and GNI) give a misleading picture. In spite of double-digit GDP growth, the percentage of Timorese people in poverty has increased to more than 50%, and we hope that the new CPI will reflect the living conditions of this disadvantaged and vulnerable majority.

Market principles often don't apply in Timor-Leste. When buyers don't know that a product is cheaper somewhere else, or if going there is too difficult, some sellers charge more. When vendors don't understand that some income is better than none, they ask the same price until a perishable product becomes unsellable. Can DNE's published price data include variation of prices among stores or localities to help reduce this lack of information?

Timor-Leste is extremely import-dependent, and the prices of imports are often very different than the prices of local products for which they substitute. DNE’s recent 2011 Trade Report shows that exports were $13 million, with $319 million in imports. The Balance of Payments is more striking – during 2011 $1,764 million left Timor-Leste while only $381 million came in (plus $3,240 million in oil income, which could drop to zero in about 12 years). Imports are mostly purchased by more affluent, urban people, and a CPI which mainly includes imported goods does not reflect most people’s lives. Is it possible to disaggregate the data, to enable comparison of changes in price differences between, for example, imported and locally produced rice? Increasing local production to substitute for imports is essential to Timor-Leste’s future, and such data would assist policy development.

Approximately half our population lives below the poverty line, largely in rural areas, and we suggest that a separate CPI be calculated to indicate their costs of living. With fewer options and less resources, they are hit harder by inflation, and it is important to measure and understand these effects.

For example, Timor-Leste imports vehicles and fuel, but almost no rural people have cars or motorcycles. They walk, occasionally using horses or public transportation when they require goods or services not available in their vicinity, such as health care or education. Rather than purchase Aqua, they walk long distances to undrinkable sources. Rather than purchase LPG or kerosene, they gather firewood. They cannot purchase many tradable items which others buy, sometimes doing without, producing for themselves or bartering in the subsistence economy. But when they really need to purchase something, inflation makes their scarce money buy less. How can the CPI reflect their reality?

Workshop attendees discussed causes of inflation, including high domestic state spending, lack of local productive capacity, foreign exchange rates, supply bottlenecks, price-fixing, importer or retailer oligopolies and limited consumer choice and knowledge. We agree that a solid understanding is essential to develop effective economic and budgetary policies, but perhaps it is asking too much of CPI to explain it. Can ABS and/or DNE undertake further in-depth analysis?

We wonder about excluding owner-occupied housing from the CPI. Construction and maintenance of homes is a significant part of most households’ expenditure. For poor people, dwellings may need more frequent repair or rebuilding, and, when purchased materials are needed, they can be expensive. However, the lack of facilities (such as plumbing, electricity, sanitation and indoor kitchens) in many poor people’s homes makes it difficult to compare them with those of more affluent people. We hope that you will find a good way to include housing costs in the CPI calculation.