A U.S. family wants to move back to India and retire. I ran the numbers.

The city they chose in India moved the number by $257,000. The life they wanted to preserve moved it by nearly $900,000.

A couple currently living in the U.S. asked me to run the numbers on moving back to India and retiring there permanently. They have two young kids and the couple plans to buy an apartment outright. They would prefer private schooling for their kids and also save funds for the university. The couple wants to travel every year, and the biggest one, they never want to work again.

I expected the city that the couple chooses to retire in to make a huge difference, but it didn’t.

Here are the numbers. Bangalore: $2.94 million. Tier 2 city: $2.71 million. Tier 3 district town: $2.68 million. Moving from Bangalore all the way to a Tier 3 district town saves about $257,000, less than 9%. Moving from a Tier 2 city to a Tier 3 town saves just $32,724.

While that was the first surprise (and very counter-intuitive), the second was much bigger. Changing how this family wants to live can save nearly $900,000, but changing where they live saves about $257,000. So, this stopped being a story about whether Bangalore is expensive.

It became a story about something much more important: The expenses that look small every year can dominate the amount you need to retire forever.

Here’s the Bangalore version of the budget, in today’s dollars:

• $300,000 for an apartment, bought outright
• $50,000 to get settled
• $10,000 per child per year for private school through age 18
• $300,000 per child for university
• $20,000 a year for family travel
• $1,000 a month for groceries, household help, utilities, and health cover
• $10,000 a year for everything else

The city changes the answer far less than people expect.

Which means the question everyone wants to debate, Where in India should I live? isn’t actually the biggest question in this model. Not even close.

Part one: Bangalore costs $2.94 million

The first thing to understand is that an annual budget and a retirement budget are two very different things. A $10,000 annual expense doesn’t cost you $10,000. If you want to fund it indefinitely, it can require hundreds of thousands of dollars of capital.

Every dollar you spend forever, costs about $37.

Start with a 3.5% withdrawal rate, that means, every $1 of annual spending requires roughly $29 invested. Now, assume Indian expenses rise about 0.5% per year faster than the inflation already embedded in the model. Your effective denominator falls to 3%. Now you’re at $33.33. Then account for the tax required to produce that dollar of spending from a taxable portfolio.

The model lands at roughly $36.87 of capital for every $1 you plan to spend annually forever. We will round it off to $37. And suddenly, some ordinary-looking expenses become huge. A $20,000 annual travel budget requires about $737,000. A $10,000 miscellaneous budget requires about $369,000. One good family trip every year requires more capital than the apartment.

The expensive things aren’t always the things with the biggest price tags. They’re the things you plan to keep buying forever.

One good family trip every year needs more capital than the apartment.

Then, there’s healthcare

Healthcare looks tiny in the starting budget. A family might pay around $2,400 a year for health insurance in their 40s. But you can’t take that number and extend it unchanged for the next 50 years because premiums rise with age and children eventually leave the family policy.

Senior policies can come with co-pays and coverage restrictions. So, I modeled healthcare separately. Started at around $2,400 a year at age 40 and added 1.5% real medical inflation. Removed the children from the family policy after 20 years and run coverage through age 95.

By age 70, the modeled premium is about $11,800 a year in today’s dollars. Capital required today: $329,823. That’s about 11% of the entire retirement number. From a line item that starts with looking like $200 a month.

That’s the problem with very early retirement. An expense that grows faster than inflation for decades can require a surprising amount of capital today.

Where the $2.94 million actually goes

The $2,937,720 breaks down roughly like this:

• $1,542,371 funds recurring non-health spending.
• $715,525 funds education.
• $329,823 funds healthcare.
• $300,000 buys the apartment.
• $50,000 gets the family established.

Let’s also think about, where does the money sit. The recurring spending is split between money needed before age 59½ and money that can eventually be funded from retirement accounts. In this model, funding the post-59½ portion from a 401(k) instead of taxable investments adds about $100,083 to the capital requirement.

Some of the retirement spending happens before age 59½, when pulling money from a 401(k) can be more complicated or costly. So that portion of the plan needs to be funded primarily from taxable investments. Spending that happens after 59½ can be funded from the 401(k), but withdrawals are generally taxed as ordinary income. That means the family needs a larger balance inside the 401(k) to produce the same amount of after-tax spending.

In this model, using the 401(k) for the post-59½ spending instead of taxable investments increases the amount of capital needed by about $100,083.

The key take-away is simple really, $1 million in a taxable account and $1 million in a 401(k) are not necessarily worth the same amount for retirement spending, because the taxes you pay when accessing the money are different. So two portfolios with the same headline value aren’t necessarily equally useful. Where your money sits matters too.

Education cannot be modeled like groceries

Living expenses and university tuition should not be modeled the same way. Groceries, utilities, and other everyday expenses continue for as long as you’re alive. You don’t know exactly how many years you’ll need to fund them, so a safe withdrawal rate makes sense.

University is different. We know approximately when those bills will arrive, so I don’t assume that money can stay invested aggressively right up until tuition is due. I use a more conservative 2.5% real return for education costs.

The family has budgeted $300,000 per child for university, or $600,000 total in today’s dollars. Because those bills are more than a decade away, the amount that needs to be invested today is lower than $600,000. When I combine university with private-school costs through age 18, the model requires about $715,525 today for the children’s entire education.

Part two: What happens if you leave Bangalore?

Now let’s move to a Tier 2 city. Think Coimbatore, Indore, Kochi, Vizag or Jaipur. “Things are 30% cheaper, so I need 30% less money.”

Except most of your life doesn’t become 30% cheaper.

School might, household help might, local restaurants and services might, but your iPhone doesn’t, your international flight doesn’t, University abroad definitely doesn’t. So, instead of applying a blanket discount, I discounted only the portion of each expense that is genuinely local.

School fees, household help and local services get the full 30% haircut. Groceries get about 10%. Healthcare falls around 14%. Utilities barely move. International travel and university don’t move at all (and rightly so).

The result: Bangalore: $2,937,720, Tier 2: $2,713,364. A 30% local cost discount reduces the total retirement requirement by only 7.6%. Why? Because roughly 61% of the capital requirement doesn’t care which Indian city you live in. Amul butter doesn’t suddenly become half-price because you left Bangalore. Neither does a Samsung refrigerator. And your flight to Europe costs roughly the same.

Part three: Tier 3 gets weird

Let’s take the same exercise one step further. Assume local costs are 50% cheaper than Bangalore. The model gives me: Tier 3: $2,680,640. That’s only $32,724 cheaper than Tier 2.

You save more on the apartment and school. But I also modeled an additional $3,000 a year for medical access in a town without the same tertiary-care infrastructure and another $2,000 for travel friction from living somewhere without equivalent airport connectivity.

Those are small annual numbers. But remember the $37 rule, permanent costs are expensive. At 0.5% real cost drift, Tier 3 saves about $32,724 over Tier 2. At 1%, the advantage falls below $10,000. At roughly 1.17%, it disappears. Above that, Tier 3 becomes more expensive.

You are trading a fixed discount on a house and school fees for a permanent subscription to transportation and healthcare access. That is why tiny changes in long-run assumptions can flip the answer.

But there is a version where the city really does matter

There’s an obvious criticism of this model. I gave this family a $20,000 international travel budget and $300,000 per child for university. Then I declared those expenses geography-independent.

Of course, Bangalore versus Tier 3 doesn’t move the total very much. So let’s change the lifestyle too. Instead of $300,000 per child for university, assume $60,000 per child for an Indian private university.

Instead of $20,000 a year of international travel, assume $6,000, mostly domestic. Now the numbers look very different. The original lifestyle requires: Bangalore: $2,937,720, District town: $2,680,640.

The localized lifestyle requires: Bangalore: $2,038,969, District town: $1,700,096.

Now the city choice is worth 16.6% instead of 8.8%, there’s an even bigger result hiding there. Changing the lifestyle saves about: $898,751. Moving from Bangalore to the district town saves: $257,080.

So localizing the lifestyle is worth about 3.5 times as much as changing the city.

Localizing university and travel saves about $898,751. That’s about 3.5× the savings from moving cities.

That’s the decision, not Bangalore versus Indore versus a district town. It is whether you are moving your address to India while keeping a dollar-priced lifestyle, or actually moving your lifestyle to India.

Don’t count on the rupee saving you

There’s another tempting argument: “The rupee keeps falling against the dollar. If my portfolio stays in dollars, India will get cheaper every year.” I wouldn’t build a retirement plan around that. A large part of long-run rupee depreciation exists alongside higher Indian inflation.

The currency moves, but prices move too. There is another risk. The cheapest places today may experience the fastest cost convergence. The cook who costs $100 a month in a district town costs that because local wages are low.

That’s also precisely why the wages are capable of rising dramatically over the next 30 or 40 years. Schools get better, and more expensive. Healthcare improves, and gets more expensive. Services become better, and get more expensive.

The geographic discount you are counting on may be the exact discount most likely to shrink over a multi-decade retirement.

So what would I actually focus on?

If I were to make a decision based on this model, I wouldn’t start by asking whether Bangalore is too expensive. I would start with the decisions that actually move the number.

1. Localize university and travel: about $899,000
2. Cut annual family travel from $20,000 to $10,000: about $369,000
3. Choose Indian university instead of US university: about $353,000
4. Move from Bangalore to a district town: about $257,000
5. Model age-adjusted healthcare instead of holding today’s premium constant: about $236,000
6. Halve school fees: about $137,000
7. Use the more tax-efficient account structure for perpetual spending: about $100,000

Lifestyle choices beat location choices. The biggest wins come from what you keep buying for decades.

That ranking is much more interesting to me than Bangalore versus Tier 2 versus Tier 3 because most of the big levers aren’t about geography but about what kind of life you’re trying to preserve for the next 50 years. A $10,000 annual reduction in travel is worth more than the difference between Bangalore and a district town. Healthcare assumptions can move the model by hundreds of thousands of dollars.

And something that looks tiny in an annual budget can become enormous once you ask the portfolio to fund it for decades. That’s probably my biggest takeaway from running this model.

People obsess over the visible expenses because they’re easy to understand. Should we live in Bangalore? Should we move to a Tier 2 city? Can we get household help for $150? Is private school really $10,000?

While those questions matter, there are better questions are hiding underneath them. Where will the kids go to university? What kind of travel do we want for the next 50 years? What happens to healthcare at 70, not 40? Which expenses are truly local? Which expenses follow you regardless of where you live? And how much of the lifestyle are we actually willing to change?

You can save $32,724 by choosing a district town instead of a Tier 2 city. You can save roughly $899,000 by changing the lifestyle itself. That’s why I wouldn’t start this retirement plan by asking where in India to live.

I’d start by asking: What kind of life are we actually trying to fund?

Because once that answer changes, the retirement number changes much faster than the pin on the map.

Model notes, they are fun, I promise

These are hypothetical, illustrative numbers, not personal financial, tax, legal or investment advice.

For the recurring expenses, I used a 3.5% withdrawal rate, then added 0.5% real cost drift because I don’t love pretending today’s prices stay perfectly behaved for the next 50 years. I also adjusted for whether the spending is coming from taxable investments or a 401(k), because the tax wrapper matters.

Education gets treated differently. College is a bill with a date attached to it, so I modeled it as a dated liability using a 2.5% real discount rate, rather than pretending the stock market will politely cooperate the year tuition is due.

Healthcare gets its own model too: age-banded premiums plus 1.5% real medical inflation through age 95. Because using the health-insurance premium of a 40-year-old all the way to age 90 would be… optimistic.

For Tier 2 and Tier 3 cities, I did not just slash the whole budget by 30% or 50%. Only costs that should actually respond to local pricing got discounted. Your household help might get cheaper. Your international flight probably doesn’t.

And for Tier 3, I added extra costs for healthcare access and travel connectivity, because cheaper housing is less exciting if every specialist appointment turns into a road trip.