Short-Term Financing: Bridging the Operating Gap

Short-term financing covers the stretch between money going out and money coming back. Sankalp Industrial Systems Limited, invented, runs a working capital line of Rs 1,00,00,00,000 at 7.60 per cent charged monthly, pledged on receivables and inventory, renewed once a year. The line pays for a 47.45 day cycle that restarts the moment it closes, and by the end of Year 5 the balance stands at Rs 2,25,00,00,000.

Underneath that sits a mismatch that almost every business runs and almost nobody names out loud. The thing being funded keeps coming back: a cycle closes and reopens the same afternoon, so the money put into it does not return for long enough to be called back. The thing doing the funding has a twelve month life and a review date attached. A requirement that never leaves is therefore carried on an arrangement somebody else re-agrees every year, and that works, year after year, until a year when it does not. Every block below comes out of holding those two facts next to each other.

What is the gap, and why will it not close?

Think about a caterer who takes on a wedding. Provisions get bought a week before anybody eats. Staff get paid on the day. The invoice goes out afterwards and settles some weeks later. For that whole stretch the caterer has spent real money and received none, and the size of the hole has nothing to do with whether the wedding was profitable. The wedding was profitable. The money simply had not arrived yet.

A manufacturer runs the identical shape at a larger size and on a longer clock. Sankalp Industrial Systems Limited, invented, buys steel and castings, holds them as stock while orders are made up, ships to customers who pay on terms, and pays its own suppliers somewhere in the middle of that. Measured out, the stretch runs 47.45 days, a figure established under the cash conversion cycle and taken here as given. For financing, the length of the gap matters less than the fact that a new one opens the moment the last one shuts.

A gap and a project part company here. A project ends. A cycle does not. The day a customer settles is the day the next batch of steel is already sitting on the floor waiting to be paid for. So the money the business puts into the cycle is not lent to the cycle and returned. The money is put in once and stays in, and only the individual rupees rotate.

Which part of working capital is never there and gone?

Walk into a cloth shop on its quietest day of the year. The shelves are not bare. There is less on them than in wedding season, but there is a layer that is simply always present, because a shop with nothing on the shelf is not a shop. The bottom layer gets paid for once and then stays paid for, forever, as long as the business exists. Individual bolts of cloth turn over. The layer does not.

Working capital has the same floor. Reading the case entity as it stands at Year 0: receivables of Rs 2,16,00,00,000, inventory of Rs 1,44,00,00,000, payables of Rs 1,80,00,00,000, and the three together give a net requirement of Rs 1,80,00,00,000. The net requirement is money the business has put into its own operations and cannot get back while it keeps operating. The part of a working capital balance that never goes away is not a loan the business is waiting to be repaid; it is capital, and it behaves like a machine that happens to be made of stock and invoices.

The difference becomes clear on asking what would release it. A machine releases cash when it is sold. The working capital floor releases cash when the business shrinks, or when the days are cut, or when it shuts. None of those is a repayment. Calling the funding behind that floor short-term therefore describes the paperwork rather than the money.

Against that floor of Rs 1,80,00,00,000, the facility supplies Rs 1,00,00,00,000, or 55.6 per cent of it. The remaining Rs 80,00,00,000 comes from the company's own capital. So even at Year 0 the business is not borrowing the whole requirement, and the borrowed share of a permanent requirement is itself permanent.

What does the arrangement actually say, line by line?

The whole arrangement comes off four lines. Each line changes the instrument in a different way, so the four are worth reading the way a lender reads them, and three of them are easy to skim past.

Compare that with the other two borrowings on the same balance sheet. A secured term loan of Rs 3,00,00,00,000 at 7.80 per cent sits on a bullet maturityOne repayment date at the end, with nothing paid off along the way. The balance sits unchanged until the day it all falls due. at the end of Year 5. Listed non-convertible debenturesDebt sold to many holders at once rather than borrowed from one lender, and it stays debt for its whole life. of Rs 2,00,00,00,000 at 8.50 per cent run out to Year 7. How those two are structured, what a lender takes against each and how their repayment dates are managed is worked through under debt financing and debt capacity. The third line is set apart from the other two not by its price or its size but by the fact that its continuation is decided again every year.

If receivables and inventory are the security, what happens when they move?

Picture a fruit seller who borrows against the crates in his cart. On a good morning the crates are full, the lender is comfortable, and the loan is easy. On the morning after a spoiled consignment the crates are worth less, so the lender will advance less, and that is precisely the morning he most needs money to restock. The pledge did not fail. The pledge behaved exactly as written. The trouble is that the thing standing behind the borrowing and the thing creating the need are one and the same.

The facility sits in that shape. Receivables of Rs 2,16,00,00,000 and inventory of Rs 1,44,00,00,000 are what the money pays for, and they are also what the lender looks to. Pledging an instrument on the asset it funds gives the borrower no spread of risk at all, because the two sides move together and they move in the wrong direction at the same time.

Follow it through. Collections slow, so more of the receivables book is old. Old receivables are worth less to whoever might have to realise them, so the amount a lender will allow against the same balance sheet can fall even though the balance sheet looks larger. Meanwhile the slowdown itself is consuming cash, so the business wants more. Capacity goes down as need goes up, and neither movement had to be anybody's fault.

A plan that treats the drawn limit as a fixed number has therefore assumed away the only genuinely interesting feature of the instrument. The limit is written down. The amount that can actually be drawn against the limit is not a written number at all. The drawable amount tracks assets that change every week.

What does charged monthly do to a quoted rate?

A rate on its own is not a cost. A rate becomes a cost only once the frequency of the charge is known. Line 2 of the terms sheet says 7.60 per cent a year and says it is charged monthly, which means the interest is applied twelve separate times inside the year rather than once at the end, and each application joins the balance the next one is calculated on.

Put on an effective annual rateOne year of a borrowing restated as a single number, once interest landing more than once inside that year is allowed for. footing, the 7.60 per cent becomes 7.8704 per cent. Monthly charging adds 27.04 basis points to the quote, and it adds them for no reason other than the calendar on which the charge falls. How a quoted figure is converted into an effective one is worked through under discounting and rates.

The reason to bother is comparison. The other two borrowings are charged on different calendars again, so their quotes are not directly comparable with this one either. Once all three are restated the ordering does not change, and that is worth knowing precisely because most people assume that frequency reshuffles a ranking. Here it does not. Restating narrows the distance between the numbers without moving anybody past anybody.

Where does this money sit against everything else the company borrows?

Same invented company, same day, four ways of getting hold of money. Three of them carry an interest line. The fourth does not. The missing interest line is exactly why it gets overlooked, and the supplier is the most expensive of the lot by a wide margin.

The supplier line is the one with no paperwork, so it needs a word of explanation. A supplier offers 2 per cent off for settling within ten days and otherwise wants the whole amount at forty five. Declining the discount is a decision to borrow the invoice for the extra thirty five days at the price of the discount forgone, and annualised simply that comes to 21.28 per cent a year. The 21.28 per cent is worked out under the supplier discount decision and arrives here as a finished number. Stretching the supplier costs 2.80 times what the facility costs, and it costs it silently, because no line in any account is labelled with it.

Note what the effective column does and does not do. Because the calendars differ, the column moves every number up by different amounts. The column reorders nothing. The blended cost of debtAll of a borrower's debt averaged into one rate, each piece weighted by how much of it there is. across the three borrowings is 8.00 per cent before tax on the quoted figures, and 6.00 per cent after tax at the 25.0 per cent rate this forecast assumes; where that blend comes from and what it is used for belongs to the cost of capital subject.

Why is the cheapest money in the business also the least comfortable?

Something is odd about that table, and it is worth sitting with. The borrowing with the shortest life, the loosest documentation and the least certainty attached to it is also the one with the lowest number on it. The low number is not a mistake and it is not generosity.

A lender prices what it is exposed to and for how long. Lending against stock and invoices, with a review every twelve months and the right to reassess, is a shorter and better watched exposure than lending against the business for five or seven years. The facility is cheap because the lender kept the option to stop, and the borrower has effectively sold that option in exchange for a lower rate.

So the rate is not the whole price. The rest of the price is paid in a form no rate can express: a dependence on a decision that gets remade every year, and it gets remade at whatever moment happens to arrive, including moments when the business would least like the question asked. There is no line in any account for that, and no plan captures it by carrying 7.60 per cent forward five times.

Notice too what cheapness does inside the business. Cheap money is used more. A treasurer comparing 7.60 with 8.50 will naturally put more of the requirement here, and the forecast does exactly that. The effect over five years is that the arrangement with the annual review ends up carrying more of the business than the arrangements without one.

What actually happens at a renewal?

A renewal is often described as an administrative event, and describing it that way is where most of the trouble starts. A renewal is a lender re-examining a borrower it is already exposed to, and reaching a fresh view on three separate things: whether to continue at all, at what size, and on what terms.

Two of those can move without the third. A lender can continue and reduce the size. A lender can continue at the same size and reprice. The renewal is not one decision with a yes and a no; it is three decisions, and only the first of them ever gets discussed as if it were the risk.

There is a household version that lands uncomfortably. Somebody who rents a home on an eleven month agreement is not homeless and has never missed a payment. Once a year, though, somebody else decides whether the arrangement continues, at what rent, and on what conditions. Nothing is wrong on any given day. The tenant carries a recurring dependence on a decision that belongs to another party, and the fact that it has always gone the same way is not information about the next time.

Why does something meant to be short-term keep getting bigger?

Here is the part readers find hardest, because it runs against an instinct. Borrowing rising year after year reads as a business in trouble. In this forecast it is the opposite: the borrowing rises because the business is growing, and nothing in it deteriorates at all.

Take a tiffin service adding twenty new customers a month. Every new customer means more rice, more oil and more containers bought before a single rupee of that customer's money arrives. The service is doing well. The service is also, every single month, further out of pocket than it was, and the better it does the further out of pocket it gets.

Put numbers on it. Every year of the forecast brings Rs 1,20,00,00,000 more revenue than the year before it, so a base of Rs 12,00,00,00,000 reaches Rs 18,00,00,00,000 by Year 5. One assumed constant sits underneath: fifteen out of every hundred rupees of revenue stays tied up in the cycle, an assumed ratio rather than one measured from any business. Apply it to the increment and the requirement gains Rs 18,00,00,000 a year, Rs 90,00,00,000 across the five, closing at Rs 2,70,00,00,000. Not one ratio in that calculation moved, and the cash requirement still rose by half.

The borrowing rises faster still. A quarter of the growth here is paid for by borrowing, so net new invested capitalWhat a year genuinely adds to the money tied up in a business, after wear and tear on what was already there. of Rs 1,00,00,00,000 a year is matched by Rs 25,00,00,000 a year of net new borrowing, and all of it lands on this one arrangement. Five years of that takes the balance from Rs 1,00,00,00,000 to Rs 2,25,00,00,000, while the term loan and the debentures do not move at all.

The two increases are not the same figure and it matters that they are not. Working capital takes on Rs 90,00,00,000 over the five years and the facility takes on Rs 1,25,00,00,000, a difference of Rs 35,00,00,000. The difference is going into fixed assets. The Rs 25,00,00,000 is a quarter of all the new capital the year puts in, and capital expenditureMoney laid out on plant, buildings and machines, which is spent once and then used for years. net of wear and tear is the larger part. Anyone treating the facility balance as a measure of working capital will be wrong, and wrong by a growing amount.

What do five years of it look like on one chart?

Read the last column, because it is the one nobody prints. At Year 0 the business funds Rs 80,00,00,000 of its own working capital from its own capital. By Year 5 that has fallen to Rs 45,00,00,000. The facility is not merely growing; it is growing into a larger share of a requirement that is itself growing, so the business becomes more dependent on it in two ways at once. The last column is the part of the working capital requirement not funded on this line, and the facility itself is doing more than working capital, which is why the two columns do not tell one story between them.

One more reading off the same forecast, and it needs a caution attached to it. Both rows below price the arrangement on its own rate rather than on the model's. The forecast itself does not charge interest that way: it charges the whole of opening gross debtThe whole of what has been borrowed, counted before any bank balance is set against it. at the blended 8.00 per cent rather than working through each borrowing separately, so those two figures illustrate this line and are not the interest line in the model.

What does a five year asset on a twelve month arrangement actually look like?

Put the two lives on the same timeline and there is nothing subtle left to argue about. The thing being funded has never been below Rs 1,80,00,00,000 and is on a path to Rs 2,70,00,00,000. The thing funding it has a life of twelve months and then has to be agreed again, five times across the same stretch.

The shape has a name, a maturity mismatch, and the name is as far as anybody can go. A mismatch names an arrangement and states no outcome. Nothing in the case entity says whether the arrangement was renewed on any of those five occasions, on what terms, or at what size.

Two things a mismatch is not: a prediction, and a claim that trouble is coming. A mismatch is a statement about where a decision sits. The continuation of the funding is not inside the business's control, and the requirement it funds is not going anywhere. Both halves of that sentence have to be true for the mismatch to exist, and on this record both are.

What does a lender see in this balance that a shareholder does not?

Three people look at the same balance and take three different things from it, and the differences are not matters of opinion.

A lender looks first at what stands behind the money and how fast that is turning. The receivables book and the stock are the assets it can look to, so their age and quality matter more than the profit and loss account does. The lender then looks at interest coverA lender checks it in one line: how many times over would a year of operating profit settle that year of interest.. At Year 0 this invented company shows operating profit of Rs 2,40,00,00,000 against an interest charge of Rs 48,00,00,000, giving cover of exactly 5.00 times. The lender is not asking whether the business is good; it is asking whether it would get its money back if the business stopped being good.

An analyst outside the business does something different. Seeing borrowing rise from Rs 1,00,00,00,000 to Rs 2,25,00,00,000 across a forecast, the first job is to work out whether that is deterioration or growth, and the only way to tell is to check whether anything in the ratios moved. Here nothing did. Days are flat, the working capital ratio is flat, margins are flat, and the borrowing still rose by Rs 1,25,00,00,000. An analyst who reports rising borrowing as a warning without running that check has reported the arithmetic of growth as a problem.

A shareholder gets the residue and therefore cares about a third thing again: whether the growth being funded earns more than it costs. The comparison belongs to the returns and cost of capital subjects. Short-term financing contributes one observation to it. The money doing the funding is the cheapest on the terms sheet and the least certain to be there.

There is a household version of all three. A shopkeeper borrowing against stock is watched by the lender for what is on the shelf, by an accountant for whether the shop is bigger than last year, and by the shopkeeper's own household for whether any of it ends up as money to live on. Same shop, same figures, three questions.

When does short-term financing stop being the right answer?

Short-term financing answers a timing problem. Money leaves before money arrives, and something has to sit in the middle. The timing problem is real, it has a clean solution, and the solution is priced accordingly.

The test for whether a timing problem remains is simple and it is not about the instrument at all. Does the balance return to where it started inside the year? On this record it does not, in any year. The balance starts at Rs 1,00,00,00,000, ends the five years at Rs 2,25,00,00,000, and never once goes back down. When the balance has no low point except the day it began, what is being funded is not timing, and the instrument answering it has stopped matching the question.

At that point the question changes rather than getting harder. The question stops being how to bridge the cycle and becomes what kind of capital ought to carry a requirement that never goes away, a question about the whole funding structure. Capital structure covers that question separately, and it is a genuinely different argument with different tests in it. The moment of change can still be dated exactly, and it arrives when the low point of the balance stops existing.