The High Cost of Living — 2

The cost of living is high in India, as I mentioned the last time.

The dominant theme around where I live in Kalyani Nagar in Pune is one of massive construction. Multistoried residential buildings, shopping centers and office complexes are sprouting with astonishing rapidity. Despite the increase in the quantity supplied of floor space, the quantity demanded is growing even faster. This is evidenced by the fact that the price per square foot of built up space is growing at an astonishing 30 percent or more per year.

Right around the corner, the going rate is now around Rs 4,000 per square foot (or around $100 per square foot.) How expensive is that? Depends on who is paying, of course. Around the construction sites, hundreds of workers toil away ceaselessly. They earn around Rs 70 a day. I figured that to them, the per square foot cost is 2 months’ worth of earnings. Thus if about 800 square foot of space is reasonable for a family of four, the average laborer on the construction site would have to work for about 1600 months, or about 133 years to just be able to afford to buy the place. Imagine having to work for over one hundred years to be able to buy an apartment.

Of course, one needs more than just floor space in a building to live. There is food, clothing, utilities, and other needs such as medical, educational, entertainment, and so on. Taken together, all of these have a cost and they add up.

Time is a great measure for all sorts of things once you have an invariant in hand. For measuring distances, the speed of light provides an invariant measure. Thus, you can say that the sun is about 8 light minutes away and the nearest galaxy to the Milky Way, the Andromeda galaxy, is about 2.5 million light years away.

Costs can be measured in terms of time, provided we have an invariant. In our case, we really don’t have an invariant because the time to earn a unit of money varies from person to person, and from region to region. However, there is already a measure of average incomes in a specific location. That is the total production of final goods and services measured in monetary terms over a year for a specific collective of people that is called the gross domestic product (GDP). GDP divided by the number of people gives you the GDP per capita.

GDP per capita is a handy “local invariant” of sorts. India’s GDP per capita is about $450 per year. In a sense, you can say that the average person earns $450 a year and that is a local invariant. Now if something costs $450, then one may say that it costs one year. The “Per capita GDP Time Equivalent of Cost” is what I am provisionally proposing, or PGTEC. (The stress is on the word “provisionally.”) The units are time: so you could measure PGTECs in hours, days, months, years, and life times.

I can now denote the cost of an 800 sq foot apartment in my neighborhood for a laborer as 200 years. Since the working life of an average person (given the life expectancy of about 65 years and a productive life of about 40 years), the cost of the apartment would be five laborer lifetimes. Astounding.

What good is the PGTEC measure? Well, for one thing, it is easy to understand and compare costs across regions. The median house in the US costs around $70,000. And the per capita GDP is around $23,000. So the average house cost in terms of PGTEC is 3 years. Compare that to 200 years in India. India is therefore about 70 times more expensive than the US when it comes to housing.

OK, I am cheating. The cost of the house in my neighborhood in Pune is not really representative in the sense that on average an 800 sq foot apartment in India would probably cost half of that. But you get the idea. The average house in India would cost in terms of PGTEC about 100 years or so. (The actual numbers I leave to bean counters; I am not one. So there.)

It makes sense to distinguish between tradeables and non-tradeables. Tradeables are those things that can be transported from one location to another. For instance, gasoline is a tradeable while haircuts and baby-sitting are not. The price of tradeables in a specific location is the same the world over once you account for transportation costs and other barriers that add to the price such as taxes.

Let’s take the cost of a tradeable such as gasoline. Gas in the US costs $3 per gallon. That is, PGTEC cost of gas is around one hour in the US. In India, gas costs about Rs 200 per gallon. Translated into PGTECs, the cost is 72 hours. Gas (or petrol as it is called in India) is 72 times more expensive in India than in the US.

Now you may have heard that a haircut is cheap in India. Let’s do the numbers. In India, a haircut costs Rs 20, or about three hours in PGTECs. In the US, it costs $15 which in PGTECs is an hour and a half. So strictly speaking, a haircut in the US is cheaper for an American compared to a haircut for an Indian in India.

Let’s go to travel. In India, the cost of a round trip air ticket between Mumbai and Delhi is Rs 14,000 average. PGTEC cost is 240 days. The cost of a round trip between NYC and San Francisco is around $400 – a distance twice that of between Mumbai and Delhi. A bit of arithmetic reveals the equivalent to the Mumbai-Delhi deal would cost an American 3 days. Roughly air travel is 80 times as costly in India than in the US.

The other night I made the mistake of going to a Pizza Hut in my neighborhood. A medium size pizza costs Rs 300 which in PGTEC terms is 4 days or 48 hours. In the US, the average medium pizza (much larger than the one in India) costs $10, which in PGTECs is only 2 hours. A Pizza hut pizza in India is therefore about 24 times more expensive in India than in the US.

I have done some back of the envelope calculations and it appears that India is many times more expensive than the US in pretty much any goods and services I measure. This should not come as a surprise to anyone who has observed that Indians can afford a lot less stuff (goods and services) than Americans can. Basically, the cost of any good is the time that you have to spend to produce an equivalent amount of stuff so that you can exchange it for that good. Indians on average produce less stuff per unit of time and therefore they can buy less stuff per unit of time as compared to Americans.

In the next posting, I will explore the implications of this measure and perhaps detail out why the PGTEC is a better measure than the PPP (purchasing power parity) measure.

Until then, goodbye, goodnight and may your god go with you.

17 thoughts on “The High Cost of Living — 2

  1. Ramnath Wednesday March 22, 2006 / 12:52 am

    Great post, Atanu.

    I have a question.

    How is it different from expressing cost as a percentage of average income (per capita GDP)?

    Let’s say, PPCI (percentage of per capita income)

    Of course, cost of some goods/services might reduce to 0.000001%

    But you can always use Per Capita Income Per Day in such cases.

    PGTEC is objective. But, what are the other benefits?


  2. Little Ram Wednesday March 22, 2006 / 11:04 am

    I think this PGTEC measure is sound,Atanu. The only caveat being that GDP per se is a one meausure of economic output and can fail to capture non-monetised services like that of housewives, etc. Also, as an example, as opposed to the US I would imagine that sons working in small business owned by their fathers in India might not get paid a real wage. So does this distort our GDP measure? It will be good to hear from you about this too.


  3. Abhijat Wednesday March 22, 2006 / 11:25 am

    Dear Atanu,

    Great fun reading this post! I am not an economist, so I lack prerequisite information for critical analysis. But a question (excuse, if naive): To what extent is the GDP sensitive to processes that determine costs/prices ? The basic concern is that a measure like PGTEC should be immune to details of underlying processes. In other words I am unable to see the invariance of GDP very clearly. If I have to RTFM, I’ll appreciate a few pointers, maybe offline, if you please.


    – Abhijat


  4. Brough Turner Wednesday March 22, 2006 / 7:20 pm

    There’s a US-only post using this same approach here:
    Another interesting parameter is the rate that things are getting better. While there is no doubt people in the US have more “stuff”, the rate of improvement, or at least people’s expectations for the rate of improvement, may be more important for personal happiness. Growth rates in India have pulled ahead. …just a thought…


  5. Parijat Wednesday March 22, 2006 / 8:01 pm

    The numbers you have used seem off significantly (call me a bean counter but its important to be one to spot HUGE errors). India’s GDP per capita is actually 736 dollars (CIA factbook). Also, a decent house for four can be had for about 8 lakhs (not 32 lakhs) – you should agree because clearly $70,000=approx 30 lakhs… so clearly you are not saying that houses in India cost more than in the US in absolute costs. That means that the cost of a house is around 25 years (not 200 years or even 100 years).

    This is a first cut response. I haven’t found time to do all the bean-counting yet… will come back with more thoughts later.


  6. Varada Thursday March 23, 2006 / 1:44 am

    Leaving all the technical jargon aside, here is a news item (Paisa, yeh kaisa?) that supports Atanu’s theory though in a somewhat different manner.
    It says 25 paise coins are not being accepted by shopkeepers and the general public inspite of being legal tender because they are considered to be of no value. Contrast this with the USA where even 1 cent is gladly accepted and given, no questions asked.


  7. Parvati Thursday March 23, 2006 / 10:50 am

    An “Aaaah” post!
    Basically, I think(maybe in a simplistic way), that if we earn more in one second then, automatically the cost of living decreases.


  8. M Thursday March 23, 2006 / 7:59 pm

    Speaking of food, I too made the mistake of going to Pizza Hut a few times(once in 6 months). The difference in the quality of pizza here and that of in the US is astonishing. Dominos even more so. The stuff here is garbage. Too much refined flour. Microscopic toppings. And pathetic base. I order once in 6 months to remind myself why I don’t eat the garbage that these places dish out. In 2002, the price of a medium pizza at Pizza Hut was about 400 Rs. That was then nearly the price of a medium pizza in the US!. Pizza Hut has reduced prices but it’s still too much for too little.
    About 6 months back, ordered a foot sub from subway. Gawd, no pickles to speak of and too many green chillies. And worst of all, nearly 280 Rs. That is barely cheaper than US subway prices. As for McDonalds, haven’t gone to any in India yet. But from what I hear, no comparison to the US at all.


  9. Deep Thursday March 23, 2006 / 10:21 pm

    This is a nice approach. I have three points to add.

    Firstly, the value of time differs across geographies, so you have to build in an exchange rate that captures these relative valuations. Once you do that, you may find that the US is more expensive than India, especially in non-traded goods.

    Secondly, your measure has the feel of a proxy for productivity. I am not surprised that India is less productive than the US.

    Thirdly, there is some information in the data that you use that you have overlooked. What matters for comparison is not only the separate prices of tradeables and non-tradeables, but also, the relative price between these two things in any one location. You will find that the price of non-tradeables in terms of tradeables will be much higher in the US than it is in India. Economic theory tells us that this is what we should expect, because the country with a higher productivity in traded goods (the US) will have its non-traded goods sector competing for labor with traded goods; high productivity in the traded goods sector will raise wages there, but the competition for labor will ensure that the non-traded goods sector inflates as well.


  10. Ramnath Friday March 24, 2006 / 8:39 pm

    Parijat, we are talking about relative per capita GDP here. So, when you took India’s per capita GDP ($735.6/1.08) from CIA Fact Book shouldnt you have taken the corresponding USA figure ($41,800) as well from there?

    And ‘decent house for four’ is subjective, what’s it – 800sqft, 1000sqft, 1200sqft?


  11. Ashish Sunday March 26, 2006 / 2:08 am

    I agree with you method of analysis only if GDP is property measured. In India I doubt if that is the case. Take the example of haircut. Based on per capital GDP of India, haircut look expensive compared to USA. But I really doubt the GDP of India includes the informal service sectors such as hair cutting saloons. So, GDP is underreported and so is the per capital GDP. Most of the poor sections of the society actually trade n such an informal sector.

    If you take the population that trades in the formal sectors (which account for reported GDP) in India and then calculate the per capital GDP, then recalculate your figures, they will be different.

    Do you really think that even average India requires to work for 3 hours just to get his hair cut done? If that is the case, then person doing the hair cut must be very rich. He can command 3 hours worth of work of other labourer for what, an 1/2 worth of work!

    You see the problem in the calculations?

    Your calculations for gasoline makes more sense. Both US and India need to import gasoline and so this comparison is much more valid. But average person in India does not own an automobile. Also, you don’t need to travel long distance in India just to get your groceries.

    The point here is that economy adjusts to the local circumstances and therefore any comparison across the countries is going to difficult to make.

    Also, remember that Indians can take advantage of cheaper imports by opening up their economy. But Americans can’t take advantage of cheaper services (like haircuts) by opening up their immigration sector. Why? Loss of political power, risk of violence and riots and other factors make it more difficult.

    So, I would bet that a country that has higher population density and freerer economy would have much better standard of life (in terms of conveniences and not necessarily in terms of physical goods).


  12. Rishikesh Shetty Sunday April 2, 2006 / 1:18 pm

    I think what you should have done is compare the average per sq foot price in India with the per capita income. Otherwise the comparison is just not fair. In the article you do assume that the average price per sq foot would be 2000. I think its way off. It has to be lesser that that. Dont you think?


  13. Manoj Guta Saturday April 8, 2006 / 10:26 am

    This is in reponse to M’s comment. First of all, the Pizza hut in india dishes out even 4 pan pizzas at 200 bucks, i.e. 4 pizzas for about US$6. Secondly, the quality of Pizza hut in US sucks and sucks big time. Next, the McD at US is considered a happy place for kallus only. The burgers there suck. Over here in India, it is so many times better. so, better to talk after experiencing yourself, rather than believing on hearsay.

    As for the author’s argument, there are too many flaws. India’s GDP is $3100 = Rs.139500 p.a. = Rs. 382 per day = Rs. 48 per hour. So, it means a haircut costs less than 30 minutes of work in india. By contrast, US per hour GDP works out to $14, while a hair cut here costs in between $20-30. So, that is worth about 1.5-2 hours of work in US, meaning 3-4 times costly compared to India. Similarly, the cost of housing in India is highly overstated. It will cost about Rs.5-6 lacs on an average for a one bed atmt in india (not talking about Delhi or Mumbai, because in such a case I should be comparing the Manhattan and LA aptmt rates, where one bed aptmt comes in the range of $500,000 and upwards). So, it will take 4-5 years for a aptmt in India, while in US it will take about 2 years. So, the housing in India is twice as costly. Moreover, the figures get accentuated against India, because of the huge urban-rural disparity. More than 60% population is still in rural areas, where the per capita GDP is only about 20% of the urban PC GDP. In case of US, the rural population is negligible and disparity less evident except in some states like Louisiana or in central parts. so, if we only look at urban purchasing power parity, Indians are no less poor cousins of their US counterparts. At $70,000 p.a. income in NY, one enjoys same standard of living as in Mumbai at Rs.700,000 p.a. That’s been my experience at least! Write back at for further discussions…


  14. Aditya Sunday April 9, 2006 / 1:48 am

    very interesting post atanu … the numbers threw me off, i was sure there had to be a big difference after bean counting, and reading some of the comments has made me more likely to believe some of the statistics mentioned therein.

    looking forward to seeing your response to the comments


  15. vishal Tuesday April 11, 2006 / 11:46 pm

    With regards to Pizza Hut, Dominos and Mc D, you guys have totally missed the point.

    These foodstuffs are the staple diet in USA and are common mans meal. So there are millions of pizzas and burgers are sold daily, driving down prizes. In India they are considered “elite” ! Hardly a thousand odd pizzas are sold a day thereby increasing prizes !

    You should compare a burger with a samosa. A commom mans fast food in both countries and then see …a samosa costs just Rs 4-5 … i.e approx 10 cents add another 10 cents for a chhota pepsi and you have a meal in 20 cents … now compare that to a burger and soda !!!


  16. Kishore Jain Saturday June 10, 2006 / 1:41 am

    Dear Sir,
    I am working with a multinational company in USA who is into manufacturing of medical devices.
    I am getting a salary of $70,000.I live in New Hampshire.
    If I relocate to India with the same company what salary should I expect.Should I expect the same1:1 or more or less.
    Your reply will be appreciated,
    Kind Regards,
    Kishore Jain


Comments are closed.