Deflating GDP

[Marry Lee]

Dear all,
I am new to deflation of variables on Stata.
Could you please help me clarify these two issues?
My country is China, it's a yearly GDP variable that I want to deflate, my base year = 1978
I found that I should be doing this:

gen real_GDP = GDP/(CPI/100)

Is my code right?

I wanted to know whether the base year should be a year that is included in my data time interval or not necessarily?
For example, I have data for 1980 untill 2000, is it ok to have the base year= 1978?

Also I wanted to know if it is possible to use a CPI data defined as last year= 100, i.e. for each year, the base year is the last year.

Finally, is it ok to deflate GDP per capita instead of GDP ?

Any remark is very appreciated,
Thank you.

[Rhys Williams]

I'm not sure I understand all that you are asking but

(1) yes, it is ok to have a base year of 1978 even if your data is in a different range, as long as you specify this in your research paper. I presume there is a sensible reason why you want 1978 to be the base year?

(2) you should deflate both GDP per capita and GDP to be in the same base year if you are using both variables. It probably won't make much sense if you compare GDPpc in 1978 prices with GDP in current prices.

I don't understand your CPI question.

Best,
Rhys

[Justin Niakamal]

A couple of additional thoughts –

I don’t know much about Chinese data, but US GDP isn’t deflated using CPI. I would search for a China GDP deflator or just use real GDP if you have access to it. A GDP deflator is a price index of all goods and services produced, not just the stuff purchased by consumers (CPI). The choice of base year is somewhat arbitrary and tends to change over time (at least for US series such as GDP, industrial production, etc.). Lastly, I would deflate nominal GDP to get a real series and then divide by population to get real per capita GDP.

[Marry Lee]

Rhys Williams thank you for your answer. In fact, 1978 is not my choice, I just found a CPI dataset with 1978 as the base year.
For the CPI data question, it consumer price index data but there is not one base year, for every observation, the base year is the previous year, is it possible to use such data to deflate?

Justin Niakamal thank you for your answer. In fact, I found a paper that converted GDP into GDP per capita then used CPI to deflate, does this make sense to you?

[Justin Niakamal]

@Justin Niakamal thank you for your answer. In fact, I found a paper that converted GDP into GDP per capita then used CPI to deflate, does this make sense to you?

Yes. You'll arrive at the same result if the correct deflator is used. But as I mentioned above, GDP is not deflated using CPI because CPI measures goods and services purchased by consumers, not goods and services produced. GDP is a measure of the value of goods and services produced. Here's an exerpt that outlines this:

The CPI and the GDP price index and implicit price deflator are alternative measures of inflation in the U.S. economy. The choice of which one to use in a given scenario likely depends on the set of goods and services in which one is interested as a measure of price change. The CPI measures price change from the perspective of an urban consumer and thus pertains to goods and services purchased out of pocket by urban consumers. The GDP price index and implicit price deflator measure price change from the perspective of domestic production of good and services and thus pertain to goods and services purchased by consumers, businesses, government, and foreigners, but not importers. In addition, the formulas used to calculate these two measures differ.

Source:
https://www.bls.gov/opub/mlr/2016/ar...ot%20importers.

[Rhys Williams]

Rhys Williams thank you for your answer. In fact, 1978 is not my choice, I just found a CPI dataset with 1978 as the base year.
For the CPI data question, it consumer price index data but there is not one base year, for every observation, the base year is the previous year, is it possible to use such data to deflate?

Justin Niakamal thank you for your answer. In fact, I found a paper that converted GDP into GDP per capita then used CPI to deflate, does this make sense to you?

Hi Mary,

Just to note, you can rebase your index to any year you like. I would suggest choosing something sensible rather than the year available from the source.

Best,
Rhys

[Marry Lee]

Hi Rhys Williams,
Can I ask you how to do that please? If what I have is values relative to 1978, how do I change the base year?
Best,

[Rhys Williams]

Hi Marry (apologies for unintentionally misspelling earlier),

If you have something like:

1977 96
1978 100
1979 105
1980 103

Then you just divide each observation by the new base year (and then multiply by 100).

So, if we want to change the base year to 1980, then we just divide each year's index number by 103 (and multiply by 100):
1977 96/103
1978 100/103
1979 105/103
1980 103/103

This gives us:
1977 93
1978 97
1979 102
1980 100

You then use these new obs in your analysis.

In terms of Stata code, something like this should do the trick

gen index1980 = (index1978/103)*100

Best,
Rhys

[Marry Lee]

Hi Marry (apologies for unintentionally misspelling earlier),

No problems, don't worry about it!

In terms of Stata code, something like this should do the trick
Code:

gen index1980 = (index1978/103)*100

Thank you so much. That's great!