Index to use depends on the analysis you want to do . If you measure the impact on supply, probably the PPI can help you more if you are looking to isolate the volatility of prices the GDP deflator is useful ( beware the deflator base year, this switches between countries) .

Dear Cristhian, 

Thank you very much. I just want to explore the real growth of trade volume after isolating price effect. And how about the trade-off between GDP deflator and CPI in this case? 

Personally, I think GDP deflator is better than CPI for trade data (although CPI is generally more popular), because GDP deflator can measure a wider range of products at a more macro level, especially beyond consumer goods. But I am not very certain if I am right.

BTW, I find that some studies directly show trade values of different years in current dollars, and even make comparison based on such "current" values. So what is the rationale behind this? Or simply for the sake of simplicity (or even just saving time)? 

Thank you very much again!

Dear Yue:

If you look at the volume of trade in a country you can use the GDP deflator and compare the actual dollars that trade volume with variale of your interest .

If you want to analyze the impact of policy on trade volume I suggest using the net present value for a period of study and considering a national rate of discount ( taken from the country) and other international ( country's own currency ) .

If you want to analyze the volume of international trade using a model of a general or partial equilibrium do not suggest using the GDP deflator and the CPI, only sets a base year and brings flows to present value . The reason is that these models are estimated world prices as a result of agricultural policies of the countries , so the price volatility does not respond only to the GDP of an economy.

Dear Cristhian,

Thank you very much again!

Good morning;

I am afraid that you need to deflate trade using specific trade deflators. Otherwise, you will miss a very important dimension: changes in (barter) terms of trade.  

When the average price of exports increases more rapidly than import prices, the economy (firms, households, administration) gains in international purchasing power. Obviously, the inverse is true when the import prices (e.g., price of oil, for non oil producers)  increases more rapidly than the price of your exports.

Finding the adequate set of price deflators should not be too cumbersome: most national accountants publish their figure in current and constant prices.  The UN Statistical Division or the IMF website should have that too. World Bank WTI and UNCTAD Statistics have also data which may interest you.

In case of needs, a quick and dirty approach is, for exports of services, to use GDP deflator; for exports of commodities, to use international price indices for agricultural, fuels, minerals, etc... (easy to find almost everywhere: IMF, World Bank, UNCATD, WTO); exports of manufacture are deflated by either GDP deflator or (better if you believe in markets) the Manufacturing Unit Value published by World Bank.  The same MUV index will be used to deflate the manufacturing and services imports, while commodity imports will be deflated by the relevant international commodity price indexes.

Obviously, because the different aggregates of the GPD identity will be deflated using different price indexes, the basic identity GDP+M = C+I+X  does not hold anymore. The difference is attributable to gains (or losses) from terms of trade.

Best,

Hubert

Dear Hubert,

Thank you very much for the detailed answers. Actually, I simply aims to analyse the real growth of exports of various product groups (at an aggregate level, SITC 2 digit) across various country groups. So I think terms of trade may not be an important aspect here (this is also why I think I can deflate the data). 

You offered a set of price indexes for deflating different types of goods. But in order to deflate the overall export values of all goods, which one do you think is the suitable price index? Is GDP deflator a good candidate?

Moreover, I find that in UNCTAD commodity price database, there are both "price index in current dollars" and "price index in constant dollars". The two indicators, of course, have different values, but their values in the base year (2000) are the same (100). So I am not certain about their difference. And which one shall I use for deflating commodity export data?

Thank you very much again.

Best

Yue

Dear Yue:

If you can access data for the country of your interest, just look at the disaggregation of national accounts in current and in constant prices.

Comparing them will give you the implicit deflator.

I suspect most SNA presentations put net trade  (X-M) on the demand side of the GDP identity: look at GDP from the demand (use) side at some international organization (UNSD, IMF or World Bank). Some international organisation also publish balance of payments in current and constant price. So, you can get the international price deflators from them.I know (I worked there many years) ECLAC does it for Latin America, perhaps ESCAP does it for Asia and ECA for Africa.

Best, HE

Many thanks again!

Hi

I am studding fish export growth in international market. Could you please recommend me a price index to bring all princess in to one base year

Thanks