Search interesting materials

Friday, October 05, 2007

Does urban India favour liberal economics?

There is a broad stereotype in the world, where it is felt that liberal views about economic policy and globalisation are held the most in the US and the UK. It is often felt that mainstream middle class urban India has a fairly socialist approach to many questions.

The Pew Institute has been running a large scale survey effort across the world. In their latest 2007 survey, their sample is roughly 45,000 people worldwide. In India, they have roughly 2,000 people, with an urban bias.

They have three key questions that measure economic liberalism, covering attitudes towards international trade, attitudes towards foreign companies and attitudes towards free markets. The results contain many surprises. As an example, in urban India, they find 89% are supportive of international trade, 73% are supportive of foreign companies and 75% are supportive of free markets.

The fraction of the sample that is supportive on these three dimensions, in 46 countries, is as follows:

Free trade Foreign firms Free markets
US 59 45 70
Canada 82 48 71
Argentina 68 39 43
Bolivia 80 49 53
Brazil 72 70 65
Chile 88 63 60
Mexico 77 65 55
Peru 81 61 47
Venezuela 79 74 72
UK 78 49 72
France 78 44 56
Germany 85 47 65
Italy 68 38 73
Spain 82 56 67
Sweden 85 53 71
Bulgaria 88 52 42
Czech Republic 80 63 59
Poland 77 60 68
Russia 82 46 53
Slovakia 83 72 53
Ukraine 91 47 66
Egypt 61 68 50
Jordan 72 59 47
Kuwait 91 68 65
Lebanon 81 64 74
Morocco 70 72 66
Palestinian Territory 69 43 66
Israel 90 69 72
Pakistan 82 39 60
Bangladesh 90 75 81
Indonesia 71 62 45
Malayasia 91 80 71
China 91 64 75
India 89 73 76
Japan 72 54 49
S. Korea 86 54 72
Ethiopia 86 70 47
Ghana 89 89 75
Ivory Coast 94 80 80
Kenya 93 82 78
Mali 86 79 76
Nigeria 85 82 79
Senegal 95 87 63
S. Africa 87 77 74
Tanzania 82 45 61
Uganda 81 73 67

I think all three measures are driven by an underlying value system that I term `the liberal worldview'. A summary statistic could be made by averaging the three. I did a principal components analysis and find that the weights on the first principal component are not that unequal - they are 0.59 for trade, 0.59 for foreign firms and 0.55 for free markets. The first principal component accounts for 61% of the variance. Using this principal components analysis, one gets this table of countries sorted by the prevalence of liberal views on economic policy:

Liberal worldview
Ivory Coast 2.42
Kenya 2.33
Ghana 2.20
Bangladesh 1.98
Senegal 1.93
Nigeria 1.83
Malayasia 1.76
Mali 1.62
India 1.58
S. Africa 1.51
China 1.29
Israel 1.28
Kuwait 0.96
Venezuela 0.73
Uganda 0.58
Lebanon 0.55
S. Korea 0.38
Chile 0.29
Sweden 0.21
Ukraine 0.13
Slovakia -0.03
Spain -0.07
Poland -0.19
Canada -0.20
Ethiopia -0.21
Brazil -0.27
Morocco -0.27
Czech Republic -0.31
Germany -0.34
UK -0.38
Mexico -0.64
Tanzania -0.83
Peru -0.93
Bulgaria -1.08
Pakistan -1.13
Russia -1.19
Bolivia -1.20
France -1.40
Italy -1.48
Palestinian Territory -1.56
Jordan -1.63
Indonesia -1.68
Japan -1.74
Egypt -1.87
US -1.96
Argentina -2.95

This does not fit well with the stereotype of liberal Anglo-Saxon economics being prevalent in the US and the UK and nowhere else. If anything, poor countries have even more liberal values when compared with rich countries.

Urban India is at rank 9 out of 46 countries. China is at rank 11. I guess both countries have seen socialism first hand and support for liberal economics is strong.

The report also shows sharp changes over the last five years. On page 16, they show support for foreign companies in India went up from 61% in 2002 to 73% in 2007, a gain of 12 percentage points. On the question People are better off in free markets, support went up from 62% in 2002 to 76% in 2007, a gain of 14 percentage points. Most interesting is page 20, where the fraction that believes that government has too much control has risen from 52% in 2002 to 71% in 2007 - a rise of 19%.

If you looked at the rhetoric of political parties in India, you wouldn't think that this was the way the people think. Of course, this is not quite how voters think. The Pew Institute's sampling in India has a 79% urban weightage.


  1. Hmm, This is an opinion poll about what the people think about *their* current governments and *their* current economy, isn't it? Comparing the %s between US and India of people who support free markets, for instance, is comparing apples to oranges, considering the current state of the respective economies. Of course, in urban india, there's be a backlash against the pervasive government meddling everywhere. But we can only compare these numbers and make statements about who is more liberal only when the Indian government has gone as far as the US goevernment has, in terms privatizing even core functions such as defence, health care and education.

    Also, not clear what people understood by these questions: Does "free trade" mean imports from China or exporting IT services to US? Does "foreign firms" mean Nokia/Ford or KFC/Coke? Does "free market" mean "Digital camera at 2000 rupees" or "rice at 40/kg"? It does not look like the questionnaire presented the context clearly.

  2. there is a disconnect between urban India's prefrence for 'free markets' and 'free trade' and their policy prefrences

    there was a survey done few months back where urban middle class overwhelmingly opposed disinvestment and supported government subsidies and government intervention in the markets

  3. it is not appropriate to draw conclusions broad conclusions based on a sample size of 2000 people in a country of billion people

  4. Anonymous,

    That's not true.

    The efficiency of an estimator depends only on the characteristics of the sample and the sample size. It does not vary with the size of the population. Assuming Messrs Pew et al did their experimental design correctly, 2000 odd people gets them to the correct range of standard errors regardless of India's population.

  5. Can't agree more with kk. Leftist tendencies are frowned upon in India, simply because the government fails to perform. The example oft-quoted by left-liberals in the US is , there wouldn't have been the 'free' interconnected Internet as we know it today had it been left to the private firms, it would have been a similar mess as the mobile and telecom networks.

    However for whatever the reasons, the urban middle class in India does have a liberal economic view, free market does tend to perform far better(though not equitably) for them in a developing country like India.

  6. just curious: if these surveys are statistically valid,how come the results are presented without providing the sampling error %ge or any confidence interval.

    using numbers like 78% people think in so-and-so manner makes it sound like some empirical truth rather than just statistical hypothesis

  7. My apologies. The documents from the Pew website say the 95% confidence interval is at roughly 2 percentage points. So when one sees large changes (e.g. 10 percentage points) there is little doubt about statistical significance.

    sqrt(p(1-p)/N), evaluated at p=.2 and N=2000, gives .0089.

  8. I like your tally of the "liberal worldview". While I doubt Ivory Coast economic view is more liberal than UK or US, the pattern is to be expected. At the moment, poorer developing countries probably see more is to be gained from free markets and free trade than richer developed countries perception that they are on the losing end of free trade.

    The perception is among the working class in both spectrums; investors, owners of capital, and economies as a whole in both spectrums tend to be winners. Hence the push for more socialized entitlement programs in developed countries under the guise of providing safety net from free trade and global competition.

  9. In India, they have roughly 2,000 people, with an urban bias.

    Pls lets not ignore the crosssection involved here.


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.