Savings vs Investing: Certainty Against the Chance of More

Saving puts money somewhere it will still be there, in an amount that is known. Investing puts money somewhere its amount is not known and could be lower. Certainty of the amount is the whole comparison. Everything else, including how quickly each becomes money and what happens when prices rise, follows from that one thing.

Almost every version of this comparison is a comparison of growth: one side slow and safe, the other faster and riskier, and the reader invited to pick. A comparison of growth requires stating what each side grows by, and on one of the two sides no such figure can be stated. A deposit can state a figure because somebody has undertaken to pay it. The other side has nobody undertaking anything, so any figure attached to it is an assumption wearing the clothes of a fact.

Saving and investing separate instead on certainty of the amount. Set the two side by side on that and they separate at once, and five further differences usually presented as separate properties turn out to be consequences of it: how fast the money can be reached, whether it can come back smaller, who has to stay solvent, what a rise in prices does to it, and which stretch of time it suits.

A comparison that assumes the reader already knows one of the two things it compares has stopped being a comparison. So each side is built up from nothing first, and only then set against the other. Every rupee belongs to one invented household, the Bhosale household, from the earlier material on money, credit and protection. Meghna Bhosale is salaried, Ashok Bhosale runs a tailoring counter, Ira Bhosale is at school, and Rs 42,770/- leaves the household in an ordinary month once the once-a-year items are spread across twelve.

What is saving, exactly?

Start away from money. A tailoring counter takes an order for eleven school uniforms, half paid at the counter and the rest due on collection. From the moment that order is written down, Ashok Bhosale knows a specific amount is coming and knows who is going to hand it over. He might still worry about whether the customer turns up. He does not have to wonder how much. The amount was fixed the day the order was taken.

SavingPlacing money where the amount that comes back is known in advance and where somebody specific has undertaken to hand it over. is that arrangement applied to money a household already has. An amount is placed somewhere, and in exchange somebody undertakes to hand back an amount that is settled before the money leaves the household's hands. Three properties define it, and if any one of them is missing then the arrangement may be perfectly sensible without being saving at all.

The first property is that the amount is stated before the money goes in, not discovered afterwards. The claim is about when the figure becomes known, not about how large it is; whether it turns out to be a good figure is a separate question. The second is that somebody is on the other side and can be named. Money that is saved is not sitting in a vault with the saver's name on it: it has been handed to an institution which is using it and has undertaken to give it back, and that undertaking is why an amount can be stated at all. The third is that getting the money back is a matter of asking rather than of finding a buyer. There may be a notice period, a form, a queue, a penalty for asking early. The amount itself was never open, so there is nothing to negotiate.

Notice what is absent from all three. Nothing says the amount is large, or that it grows at all, or that it keeps up with anything. Saving is a statement about certainty and about nothing else, and reading anything more into the word is where most of the trouble on this subject begins.

What is investing, exactly?

Now the other side, built from nothing, with no reference to the first.

Take the same tailoring counter, and change the transaction. Instead of taking an order, Ashok Bhosale puts money into a second sewing machine. Nobody has undertaken to give him anything for it. He has bought a thing that will produce, month after month, as long as there are orders and as long as it runs. When he eventually sells it, whatever he gets will be whatever somebody is willing to pay that day. Nobody can tell him that figure now. He may end up with more than he spent, or with less, and no document anywhere settles it.

InvestingPlacing money into something that produces while it is held and whose final amount is set by what somebody will pay for it, rather than by anybody's undertaking. is that arrangement. Money is placed into a thing that produces something while it is held, and what eventually comes back has two parts: whatever it produced along the way, and whatever somebody will pay for it on the day it is sold. Neither part is settled in advance and neither is owed by anybody. That is not a defect in the arrangement; it is the arrangement.

The defining feature of investing is that no party has undertaken to return any particular amount, so the amount is discovered at the end rather than stated at the beginning. Where a return actually comes from is covered separately. The consequence is what matters here. Production is uncertain and tomorrow's buyers set tomorrow's price, so the closing amount is not a fact until it happens.

Three properties define investing, mirroring the three above one for one. The amount is discovered rather than stated. No named party has undertaken to pay it, so there is no obligation to enforce. And turning the holding into money requires somebody willing to buy at a price the holder accepts. The exit is a transaction rather than a request.

People shrink this down to the word risk. The word is not wrong, but two separate things sit inside it: the holding might produce less than hoped, and the price other people put on it can move for reasons unconnected to what it produced this year. Both feed into one place. Everything the word risk is doing on this subject reduces to a single sentence: the closing amount is not known in advance.

Why is certainty of the amount the difference everything else follows from?

There is only one difference between saving and investing, and the five other differences people list are consequences of it. A reader who knows the answer to the certainty question can derive the rest without being told them, and that is the test of a real root rather than an item on a list.

Take two of them. If somebody has undertaken to pay a stated amount, a named party is on the other side, and who has to stay solvent is answered at once. If nobody has undertaken anything, solvency is replaced by a different question: who will buy this, and at what price. And if the amount is fixed, a rise in prices during the term eats into what that fixed amount will buy, and no clause in the arrangement responds. If the amount is not fixed, it can move, and whether it moves enough is exactly what nobody can state.

Every one of the five criteria below is derived from the certainty answer rather than observed separately. Deriving them is what makes certainty the root and not just the first item on a list. The picture that follows draws that derivation as a shape.

Criterion one: is the amount known before the fact?

Certainty of the amountWhether the figure that comes back is known before the money goes in, or only becomes known at the end. is the first criterion and the parent of the other five. In saving the answer is yes, on the first day. In investing it is no, and stays no until the holding is sold. There is no middle position here, and that is what makes the criterion so useful: either an amount has been stated and undertaken by somebody, or it has not.

Certainty means the number is settled. Certainty does not mean the number is large, and it does not mean the money will definitely arrive. Certainty of the amount and certainty of arrival are different things, and this criterion covers only the first. The second is criterion four.

The most common way this goes wrong is that a stated amount is presented for something whose amount is not actually stated. If the figure on offer carries a footnote explaining the assumption that produced it, then it is arithmetic somebody did, not an amount anybody undertook to pay. Arithmetic on an assumption is a reasonable thing to look at. Arithmetic is not a stated amount, and it does not put the arrangement on the saving side.

Criterion two: can what comes back be less than what went in?

Getting back less than went in is what people usually call risk, and it drops straight out of criterion one. A stated amount cannot be lower than itself. A discovered one lands wherever it lands, and one of the places it can land is below what went in.

Capital lossGetting back less money than went in, measured in plain rupees rather than in what those rupees buy. is therefore available on one side of this comparison and not on the other. A household that puts Rs 20,000/- into a saving arrangement has at least Rs 20,000/- of stated amount to point at. A household that puts Rs 20,000/- into a holding whose price other people set may have less than Rs 20,000/- when it comes to sell, and no amount of patience, care or research removes the possibility.

Two qualifications cut against the tidy version. A capital loss is not a loss until the holding is sold, and a price that has moved down is not money that has gone. And the absence of capital loss on the saving side is a statement about rupees and not about what those rupees buy. Hold that second one: criterion five is built on it, and it is where saving is genuinely weak.

Daniel Kahneman and Amos Tversky showed that people weigh a loss more heavily than a gain of the same size, and separately that people are overconfident about their own judgement and treat a recent run of outcomes as evidence about the future. The first makes the possibility of capital loss feel larger than any arithmetic would make it. The second makes it feel smaller, whenever the recent past has been kind. A reader can be making both errors in the same week.

Criterion three: how quickly does each become money that can be spent?

ReachabilityHow quickly a holding turns into money that can actually be spent, counted in the time the need allows rather than in what the holding is worth. is the third criterion, and it also follows from the first. Where somebody has undertaken to pay a stated amount, converting the holding into money is a request, subject to whatever notice the arrangement carries. Where no one has undertaken anything, it is a sale, and a sale needs a counterparty who wants the holding, today, at a price the holder will take.

A great deal of what sits on the investing side can be sold inside a normal working day in an organised market, so the gap in speed is smaller than it used to be. The gap never disappears, though: on the saving side the amount received is the amount that was stated, and on the investing side it is the price on the day the money was needed. Being able to sell quickly and being able to sell at a known amount are different things, and only one of them can be arranged in advance.

The Bhosale household's nearest goal is 26 months away. Reachability there is not a question about how many days a sale takes. The real question is whether the money will be the size it needs to be in the month it is needed, and criterion one already settled it.

Criterion four: who has to be good for the money?

Criterion four punctures the idea that one side of this comparison is free of every risk. Saving does not remove risk; it changes the risk from the size of the amount to the solvency of whoever owes it. Somebody has to be good for a stated amount, or the statement is worth nothing.

Institution riskThe chance that the party holding the money and owing it back is unable to pay when the time comes. is the name for that, and it is not a concern invented to make the comparison look balanced. Institution risk is the reason arrangements exist in most countries to protect deposits held with banks up to some level, and why banks are supervised rather than left alone. A stated amount is a promise, and a promise is as good as the party making it.

On the investing side the question changes shape rather than going away. Nobody has undertaken to pay anything, so there is no promise to be broken; what replaces it is whether the thing keeps producing and whether somebody will want it later. Both sides depend on somebody else, and on different somebodies, once again because of criterion one.

Criterion five: what does each one do about prices rising?

Criterion five is where the comparison turns around, and it is the reason neither side can be described as the safe one. Everything above has been about the number of rupees. Criterion five is about what those rupees buy.

Purchasing powerWhat an amount of money can actually buy. Prices rising reduces it even when the number of rupees has not changed at all. is what a household actually cares about. Nobody eats rupees. A stated amount protects the rupee figure completely and does nothing about the price of a school uniform, a gas cylinder or a bus fare on the day the money comes back.

A saving arrangement protects the amount and does nothing whatever about what the amount buys, and that is not a criticism of it but a description of what it was built to do. The certainty runs out at the boundary of the number itself.

The investing side does something less definite. Holdings there are claims on things that produce, and the prices of the things produced can move when other prices move. The amount therefore has a way of responding that a stated amount does not have. A way, not a mechanism anybody has undertaken; whether it responds enough over any particular stretch is precisely the figure nobody can honestly state.

So the honest version of criterion five is this. One side is certain in rupees and carries no defence against what those rupees buy. The other is uncertain in rupees and has a way of moving that the first does not, with no undertaking that it will. The two sides carry different exposures rather than ranking against each other. Which one bites depends on how long the money sits still, and that is criterion six.

Criterion six: which stretch of time does each one suit?

HorizonHow long before the money is actually needed. It is a date, not a preference, and it is what makes one criterion bind harder than another. is the sixth criterion and it is entirely derived from the five above it. Nothing new is introduced: the five simply change how hard they press depending on how far away the date is.

When the date is near, certainty of the amount presses hardest. There is no time for anything that falls to come back, and no time to make up a shortfall from a month already spent. Reachability presses hard for the same reason, and so does the chance of getting back less than went in. The price of things rising presses very lightly. Prices do not move far in twenty-six months, and the money is about to be spent. When the date is far away, the pressure moves: a fall has time to be something other than final, reachability barely matters when nothing is going to be reached for two decades, and what the money will buy at the end becomes the dominant question. Which side of this comparison produces more was never knowable, so the horizon cannot change it; what the horizon changes is which of the six criteria is doing the binding.

Notice the one criterion that does not move. Who has to be good for the money presses exactly as hard at one month as at thirty years. A promise is only as good as whoever made it, and time does not soften that.

Why can the two not be compared on which one grows more?

Comparing growth requires two figures. On the saving side a figure exists, put there by somebody who has undertaken to pay it. On the investing side no one has undertaken anything, so no figure exists, and any number offered there is one of three things: a record of the past, an average of what happened to a group of things in the past, or somebody's assumption about the future. None is an amount anybody is bound to. Setting a stated obligation beside somebody's assumption and calling the result a comparison of growth puts a fact and a guess in the same column and invites the reader to treat them as the same kind of thing.

Two numbers in a table, side by side, in the same typeface, with the same decimal places, look like two measurements of the same kind. One is a contractual amount. The other is the output of a calculation whose input was chosen by whoever built the table. The reader cannot see that difference in the table, and that is precisely why the table is built that way.

So the two are compared here on the property that is checkable on the day of looking. Is the amount stated, and if so, by whom. The document answers that in a minute, with no forecast required and no number invented to fill a cell.

Can a household need both at once?

Yes, and that is the ordinary case rather than the clever one. The two answer different questions, and most households have more than one question open at a time.

Stop thinking about the money and think about the dates. A school fee falls in June. A goal sits 26 months away. Somebody stops working in twenty-five years. Three dates are not three amounts competing for one pot; they are three different questions, and criterion six says they press on different criteria. The June fee needs an amount that is certain and reachable, and nothing about the price of things in twenty years bears on it. The twenty-five year question is the opposite in every particular.

Saving and investing are not two answers to one question that a household has to choose between; they are answers to two different questions, and a household with two horizons has both questions open. A ranking would require one question, and there is more than one.

Having both questions open does not mean both have to be answered at once, or in any order, or at all. A household whose base is not secure has a third question sitting in front of both, and resilience is where that third question is settled. The section below shows what that looks like on one household's own sheet.

Which one is a household actually doing, whatever it is called?

Names are the least reliable thing on this subject. A thing with the word saving in its name may have no stated amount anywhere in it, and a thing described as an investment may turn out to be a stated amount owed by a named party. The name is chosen by whoever is selling, to be attractive rather than accurate.

So use the test instead. Is the amount that comes back known before the money goes in, and if it is, who has undertaken to pay it? If both halves have an answer, this is saving, whatever it is called. If either half has no answer, it is not, whatever it is called.

Three things make the test work. The test is answerable from the document rather than from a conversation. The test requires no view about the future. And it fails safe: if the stated amount and the party who undertook it cannot be found, that absence is itself the answer. An amount that cannot be found stated in writing is not a stated amount, and the difficulty of finding it is the finding.

Where does the Bhosale household sit on this comparison?

Entirely on one side, and the sheet is worth walking rather than asserting. The Bhosale household holds Rs 3,67,887/- in all at the end of year two. Sort those holdings by the test above, not by what they are called.

The gold is the interesting case. Gold is the only thing the household holds whose amount is set by other people in a market, and that makes it feel like the investing side. Gold produces nothing while it is held, though: no rent, no share of earnings, nothing coming out of it in any month. A market-priced holding that produces nothing is a third category, and calling it investing would blur the definition just built.

Then the zero. The Bhosale household holds Rs 0/- on the investing side of this comparison, and that is a correct position for this household rather than a gap in it. The reason is not a view about markets; the household does not have one. The reason is two figures established earlier: the buffer covers 0.73 months of the Rs 42,770/- that leaves in an ordinary month, and Rs 71,594/- is owed. Resilience and protection come before investing precisely because of households in this position, and there are a great many of them.

Put the certain side against the month. Rs 1,89,887/- is 4.44 months of everything the household spends. Only Rs 41,887/- of it is reachable the same day, and that is 0.98 months; the rest arrives on its own timetable. So a household can hold nearly four and a half months of stated amounts and still have a buffer account covering 0.73 months, and both sentences are true at once. Criterion three is doing its work inside one side of the comparison rather than between the two.

How does a lender read the difference between the two?

The cleanest proof that certainty of the amount is the real difference is that people with their own money at stake price it every day, without ever using the words saving or investing.

Where the holding a lender is asked to lend against is a stated amount owed by a named institution, the lender is looking at a figure that will not move. The lender can lend close to that figure, and nothing between today and repayment changes the size of what stands behind the loan. Where the amount is something other people set daily, everything changes: the lender lends less than today's price, keeps a cushion sized by how far the price could move against it, watches the price, and reserves the right to ask for more when it falls. The size of that cushion is the price the lender puts on the absence of a stated amount. Certainty of the amount is being priced there, in somebody else's money.

An analyst reading a balance sheet does the same sorting. Contractual holdings get read as amounts; holdings whose amount is a price on a date get read as prices, with the date attached. The same holding will carry a different figure next quarter without anybody doing anything. Where a set of accounts adds the two together without noting the distinction, that is itself the finding.

And a household does it too, usually without noticing. When Meghna Bhosale decides which pot Ira Bhosale's school fee comes out of in June, she is not comparing growth. She is answering criterion one and criterion three about a date eight weeks away. The practical use of the distinction is exactly that: not choosing a side, but knowing which question the date ahead is asking.

Four horizons mark the corners of that range, and they read as follows. At one month, whether the amount is known, whether it can be lower and how fast it becomes money all press at full force, and what a rise in prices does is silent. At 26 months, the Bhosale household's own nearest goal, those same three still bind hardest and the price question is still quiet. At about seven years the two poles press about equally. At thirty years the price question is doing almost all the work, and of the three near questions two have gone quiet while the third has receded without vanishing. At every one of those settings, who has to be good for the money presses exactly as hard as it did at the first.

Across its rows the table puts the shape of the whole comparison in one place: the horizon never settles which side produces more, it only settles which questions a household has to be able to answer. The claim is narrower than the one usually made on this subject, and it is the widest one that can be made without stating a figure nobody can honestly supply.

The Bhosale household, Meghna Bhosale, Ashok Bhosale and Ira Bhosale are invented.
Educational material. Not advice on any investment, tax, budget or market position.