FDI and FPI: Two Very Different Kinds of Foreign Money

Foreign direct investment (FDI) buys something that stays: a plant, a warehouse, a lasting stake in a business. Foreign portfolio investment (FPI) buys a claim that can be sold tomorrow. Money is money, so on the day it arrives the two look identical. The two are separated by how fast each can be taken back out, and every other difference between them follows from that one.

Two things already established carry the weight underneath what follows. The first is that money crosses a border for a return, and that the shape the money takes on arrival decides how quickly it can turn round and go home again. The second is that a price is set where a willing buyer meets a willing seller. A currency is a thing people buy and sell in a market like any other thing, so the rule holds for a currency exactly as it holds for onions in a mandi.

One property and one consequence follow from those two. The property is speed of exit. The consequence is that a single total, added up across kinds that move at completely different speeds, can sit perfectly still while the things inside it run in opposite directions.

What actually separates the two kinds of foreign money?

The difficulty is easier to feel in a street than in a statistic, so start away from economics. Suppose a baker runs a small bakery and two people put money into it on the same Tuesday morning.

The first buys the oven. Not a share of the oven, the oven itself, installed in the back room, wired in, hot by Thursday. The second lends the baker money against a slip of paper that says the baker will repay, and that slip can be handed to a neighbour for cash at any time.

Both handed over the same amount. Both did it on the same morning. A different question follows, and it is the only one that matters here: what does each of them have to do to get their money back out?

The second person walks down the street, finds a neighbour who wants the slip, and is out by lunchtime. The first person has to find somebody who wants an oven, agree what an oven that has been used for two years is worth, arrange to move it, and wait. The oven has not become a worse thing to buy. The oven has simply become a thing that takes time to unbuy.

Speed of exit is the entire separation between the two kinds of foreign money, and it is a property of the shape the money took, not of the amount, the sector or the passport of the person who sent it.

The property matters more than it sounds, and the reason is practical. A reader who learns the two kinds as a list of examples has a list, and a list only helps with cases already on it. Hand that reader a case they have not met before, foreign money buying a half share in a warehousing venture with a fifteen year agreement, and they have nothing to reason with. A reader who holds the property asks one question instead: to get this money out, does somebody place an order, or does somebody have to sell the thing itself? The answer sorts every case, including the ones nobody has met yet.

What is foreign direct investment, and why is it slow to leave?

Foreign direct investment is money from outside a country that buys a lasting interest inside it. Usually that means one of two things. Either it buys a productive assetSomething bought in order to make something else with it: a machine, a plant, a warehouse, a fleet of vans. It earns its keep by being used rather than by being sold on., a plant, a warehouse, a line of machines, a cold chain. Or it buys a stake in a business large enough that the buyer has a hand on how the business is run, sits in the room where decisions get made, and is treated by everybody involved as a part owner rather than a passing holder.

In the Republic of Sankhya, an invented economy, direct investment for the period ran at plus Rs 22,000 crore. Every rupee of it went into things: a cement grinding unit near the northern border, two food processing lines, a stake in a logistics venture. All of it is arriving money and all of it is now sitting in the form of something physical or something contractual with a long life.

Now ask the exit question. How does a direct investor in that cement grinding unit manage the repatriationTaking money that was earned or held inside one country back to the country the investor came from. It needs converting out of the local currency first, whatever the money was doing while it was there. of that money back to Marut, the country the investor came from?

The investor cannot place an order. There is no screen anywhere on which a half share of a cement grinding unit is quoted. The investor must find somebody who wants a cement grinding unit, or a share of one. A price has to be agreed with that somebody, and a price for a specific industrial asset is worked out by negotiation over weeks rather than read off a board. There are approvals to clear, a transfer to register, and money to convert back before it can go home. Only when the last of those is done has the money actually left.

Nothing about that slowness comes from the investor's intention: it comes from what was bought, and it would be just as slow if the investor woke up on the first morning wanting out.

The usual telling of this gets it backwards, so the correction is worth holding on to. Common tellings have direct investment as patient money and portfolio investment as impatient money, as though the difference lived in the temperament of the people sending it. It does not. A nervous investor placed into a cement grinding unit is still a nervous investor who is going to take months to leave. A serenely long term investor holding a listed bond can be gone by Thursday afternoon on a change of mind. The speed belongs to the asset, not to the owner of it.

What is foreign portfolio investment, and why is its exit so fast?

Foreign portfolio investment is money from outside a country that buys a claimA right held against somebody else, either to be paid or to a share of what is left over. A deposit is a claim on a bank; a bond is a claim on a borrower; a share is a claim on whatever a business has after its debts. which already trades in a market. Shares in listed companies, government and corporate bonds, units in a pooled vehicle. The buyer is one holder among thousands and has no hand on how anything is run.

Where it can be sold matters more than what was bought. A traded claim has a secondary marketThe place where a holding that already exists changes hands between two investors. The money passes from one holder to another and none of it reaches the business that issued the claim in the first place. standing behind it. Somebody in that market is already willing to take the other side. To leave, the portfolio investor does not go looking for a buyer. The buyer is already there.

So the exit is one step where the direct investor had four. Place the order, the market matches it, convert the proceeds, done. And because it is one step, the whole exit runs at the speed of that one step. The speed of that step is set by the liquidity of a marketHow readily a holding can be turned into cash in that market without the seller having to move the price a long way to find somebody on the other side. It is a property of the market, not of the seller. rather than by anything about the investor.

The clause about market liquidity is doing real work. The portfolio exit is as fast as the market is liquid and no faster, so a thinly traded claim in a quiet market is a slow exit even though it is a portfolio holding. The property is still speed of exit; it is just that the market, and not the asset, is what sets the speed on this side.

One claim about portfolio money is got wrong more often than any other. Nothing about the speed makes portfolio money worse money. Portfolio money is not hot, not flighty, not a lower grade of capital. It is differently shaped capital. A traded market needs holders willing to buy and sell, or there is no market and nothing gets priced at all. The speed that lets portfolio money leave on Thursday is the same speed that let it arrive on Tuesday, and an economy that wanted only the slow kind would be an economy with no traded market to speak of.

What is currency appreciation, and how does foreign money reach a currency?

There is a step between money leaving Marut and money buying anything in Sankhya, and it is easy to walk straight past. A cement grinding unit in Sankhya is priced in Sankhya rupees and is paid for in Sankhya rupees. The seller of the land, the contractor, the people who install the machines: every one of them wants to be paid in the currency they buy their vegetables in. So the Marut money cannot buy anything at all until it has been swapped for Sankhya rupees.

Which means every rupee of foreign investment, of either kind, arrives at a currency market first and only reaches the real asset second. And at that currency market, the foreign investor is a buyer of Sankhya rupees.

An inflow of foreign money is therefore not merely a payment for something: it is, first and unavoidably, a demand for the currency itself.

Now put that next to the thing already established about any market: a price is where willing buyers meet willing sellers, and more buyers arriving with the same sellers standing there pushes the price up. A currency is one more thing bought and sold, so a currency behaves the same way. More demand for Sankhya rupees, other things unchanged, lifts what a Sankhya rupee costs, and what one currency costs in another is its exchange rateThe price of one currency stated in another one. The market that sets it, and everything that pushes it about, is covered separately..

Currency appreciation is that lift: a currency appreciates when it grows dearer measured against some other currency, and one Sankhya rupee then fetches more Marut than the same rupee fetched last month. Appreciation is the whole of that definition, and it is worth learning as a definition rather than as half of a pair. A currency appreciates. The sentence is complete as it stands, a statement about one currency measured against another.

The move in the other direction has its own name, its own arithmetic, and one genuinely surprising property: the same single event produces two different percentages depending on which of the two currencies sits on top. Depreciation has a full treatment of its own, and appreciation compared with depreciation is covered separately. The route is what counts at this point: foreign money must be converted before it buys anything domestic, conversion is demand for the currency, and demand lifts the price.

Can the two kinds move in opposite directions at the same time?

They can, and they often do. Here is the Sankhya capital accountThe part of a country's external record that tracks claims crossing the border rather than goods and services crossing it. What the whole account contains and how it is put together is set out separately. for the period, taken apart line by line. Every figure holds in whole Rs crore.

Read the total on its own and it says one thing: plus Rs 19,000 crore, a surplus, foreign money on balance came in. Read the lines and it says something the total cannot. Portfolio money went the other way, at minus Rs 6,000 crore, and it went the other way during exactly the period when the account was in surplus.

The Sankhya capital account is in surplus at plus Rs 19,000 crore while portfolio money is leaving at minus Rs 6,000 crore, and a reader watching only the total would have seen nothing happen at all.

A net figureOne number left after the movements in both directions over a period have been set against each other. A net of nothing can sit on top of an enormous amount of traffic in both directions. conceals the gross traffic, so the gross traffic is worth saying out loud. Rs 25,000 crore arrived across the direct and borrowing lines. Rs 6,000 crore departed on the portfolio line. Rs 25,000 crore in less Rs 6,000 crore out is the plus Rs 19,000 crore net. The net is a difference between two real movements, not a description of one.

And the two movements do not mean the same thing about the economy underneath. The Rs 22,000 crore that arrived is in grinding units and processing lines and cannot go anywhere quickly. The Rs 6,000 crore that left could leave because it was the kind that leaves by order. One sentence about what happened to Sankhya's exposure over that period would be that the country ended the period with more money in it and less of that money able to move quickly. The total says the first half and is silent on the second.

The same headline over two very different economies

Now take a second economy, and give it a capital account that reports the identical total. Direct investment plus Rs 13,000 crore, portfolio investment plus Rs 3,000 crore, net external borrowing plus Rs 3,000 crore. Add them: Rs 19,000 crore. Same headline, to the last crore.

Check the last row rather than taking it. In Sankhya, arrivals were Rs 22,000 crore of direct money and Rs 3,000 crore of borrowing, Rs 25,000 crore in total, and not one crore of it arrived in a traded claim. The traded claims were being sold. Nil out of Rs 25,000 crore is 0.00 per cent. In the mirror economy, arrivals were Rs 19,000 crore and Rs 3,000 crore of that came in traded claims: Rs 3,000 crore divided by Rs 19,000 crore is 15.79 per cent.

Two economies, one identical headline of plus Rs 19,000 crore, and a completely different answer on how much of what arrived could turn round and go out again by order. Neither economy is the better one. The two are simply differently made, and the headline they share cannot tell them apart.

What does the split between the two kinds not reveal?

Quite a lot, and it is worth listing because the split gets asked to carry things it was never built to carry.

The split does not say which money is more useful. A traded market that prices claims properly is a thing an economy needs, and so is a cold chain. The split has no opinion between them.

The split does not say whether either kind will actually stay. Direct money can be sold, slowly, and sometimes is. Portfolio money can sit untouched for a decade and often does. The split records how fast each kind could move, not how fast it will.

The split does not say whether the assets bought were worth buying. A grinding unit built where there is no limestone is a direct investment and a bad one. The split records the shape the money took, and says nothing about the judgement behind it.

The split says nothing about the person who sent it. The word direct describes the interest acquired, not the character of the acquirer.

The split between the two kinds is about speed of exit and it is about nothing else, so reading it as a ranking of quality is a category mistake rather than a difference of opinion.

The same mistake has a twin elsewhere and naming the kinship helps it stick. Government spending gets split into the part that buys lasting things and the part that pays for running costs, and readers reliably read that split as a score, with the lasting half good and the running half wasteful. The spending split is not a score there either. It is a statement about what kind of thing the money turned into, and a badly chosen bridge is capital spending while a teacher's salary is not. Same structure, same error, two different subjects. A split that records the form of a thing is not a verdict on the thing.

How does an analyst actually use the two lines?

Not by looking at one period, which is the first thing to say, because one period of anything is a photograph and the thing being watched is a change of shape.

The practice is dull and effective. The two lines are pulled out separately for several periods running and set side by side, not added. Then one question is asked of the pair: has the composition of what is arriving been moving, and in which direction? A total that has been flat for six periods can sit on top of direct money steadily giving way to portfolio money, period after period, and the flat total is exactly what hides it. One kind quietly replacing the other underneath a total that has not moved is the change worth catching. The replacement changes how the economy behaves when something goes wrong, even though nothing in the headline has changed at all.

Why does it change behaviour? Because the two kinds respond to trouble on different timescales. If conditions turn, the holder of a traded claim can act on that view within the day, and many holders acting on the same view at once is a large movement over a short window. The holder of a grinding unit can form exactly the same view at exactly the same moment and still be four steps away from acting on it. The country has not changed. Its capacity to see money leave quickly has.

A household version of the same reading sits closer to home. Two households both take in Rs 80,000 a month. One takes it all from a single salary. The other takes half from a salary and half from rent on a small shop and interest on a deposit. The monthly total is identical and a lender looking only at the total learns nothing about the difference. The difference lives in the composition, and it only shows up when something goes wrong. Reading a capital account by its total is the same reading, at the scale of a country.

The analyst's output is a description of a shape and how it has been changing, not a view on whether Sankhya should want more of one kind or less of the other. Any decision about it belongs somewhere else entirely.

Sankhya and Marut are invented.
Educational material. Not advice on any investment, tax, budget or market position.