How Business Cycles Affect Asset Classes

A business cycle reaches different kinds of holding by different routes, so one set of conditions can help one and hurt another at the same time. A claim on a business's profits moves with those profits. A fixed rupee stream is worth less when prices rise. Both are mechanisms, not rules.

The last sentence of that answer governs everything below it. A route of influence can be traced from a phase to a holding, and the tracing stops where the route stops. Putting mechanisms together into an actual position is a separate craft with its own arithmetic, its own constraints and its own way of going wrong, and it is taught where portfolio construction and allocation are taught. What to hold in a downturn belongs there.

The mechanisms below rest on material already covered, and none of it is re-derived here. The four phases and the way the Republic of Sankhya moves through them are established: expansion into a peak, two years of contraction with the troughThe lowest point output reaches before it starts rising again. Named and placed on the path where the four phases of a cycle are taught. in year 7, then recovery in year 8. Also established is the output gapHow far actual output sits above or below an estimate of the level an economy could sustain. The estimate is worked out rather than observed, and gets revised. and where it sits in each of those years. Inflation and the way a price index is put together are taught where inflation and its measurement are covered. How revenues and costs behave across a cycle for different kinds of business is taught where cyclical sectorsParts of an economy whose sales rise and fall sharply with the cycle, as against defensive parts whose sales hold up because the purchase is hard to postpone. and defensive ones are compared. All of those terms are used in what follows.

Why does a business cycle reach a holding at all?

One question is rarely asked out loud, and the answer to it carries everything else. A business cycle is a pattern in output and prices. A holding is a slip of paper, or an entry in a record, or a physical object sitting in a locker. No wire runs from the pattern to the slip of paper. So how does one reach the other?

A cycle reaches a holding because a holding is a claim on something, and a phase reaches the something rather than the claim. Whatever the cycle does to the underlying thing is what it does to the claim on that thing. If the underlying thing is the profit a business makes, the phase reaches the claim by way of the profit. If the underlying thing is a fixed number of rupees arriving on fixed dates, the phase reaches the claim by way of what those rupees will buy. If there is no underlying thing at all, the phase has nothing to travel along, and no route exists to describe.

Here is the everyday version. A household lets out the ground floor of its house to a tea stall and, separately, has lent Rs 2,00,000/- to a cousin who repays a fixed amount every month for three years. Both of those are claims. The rent was really a claim on the stall's trade, so when the offices nearby cut shifts and footfall drops, the tea stall's takings fall and the rent gets renegotiated downward. The cousin's repayment was never a claim on the stall, so it does not move at all. The repayment is a claim on a fixed number of rupees. Same street, same year, same downturn, two claims that behave nothing like each other, and the reason is not that one is better than the other. The reason is that they are claims on different things.

What are the three kinds of underlying thing?

Sort holdings by what sits in the middle box of that chain, not by what they are called, and three groups fall out. The first group is a claim on the profits a business makes. The second is a claim on a fixed number of rupees arriving on fixed dates. The third is a holding that produces no cash at all, where the middle box is empty. The three groups are not a ranking and not a menu, and the only reason to separate them is that the route from a phase to each one is a different route.

The sorting throws a good deal away. Sorting by underlying thing ignores risk, past record, ease of sale and suitability. Each of those is a real question, and each is answered elsewhere. One question only is asked of each group: when output moves and prices move, what carries the movement across, and how far does it get?

What happens to a claim on a business's profits?

A claim on profits inherits whatever profits do, and profits do something particular in a cycle: they move further than output moves. The widening is not a market effect and has nothing to do with sentiment. The widening is arithmetic sitting inside the accounts of the business, and it comes from the shape of the cost base.

Think of a wedding caterer. Some costs move with every booking taken: ingredients, hired serving staff, fuel. Some costs do not move at all across a bad season: the rent on the kitchen, the loan repayment on the refrigeration units, the two permanent cooks who are kept on because letting them go means losing them for good. When bookings fall five per cent, the first kind of cost falls with them and the second kind sits there unchanged. Profit is what is left after both, so profit absorbs the whole shortfall on a base that has already been shrunk. A cost base that will not fall with revenue is what turns a small movement in output into a larger movement in profit, and this amplification is called operating leverageThe extent to which a business carries costs that do not fall when sales fall. Taught in full where cost structure and profit measurement are covered., taught in full where a business's cost structure is covered.

So the route runs output, then revenue, then profit, and the movement grows at the last step. A claim on those profits is therefore a claim on something that already swings wider than the economy did. Work it on Sankhya. Take Sarani Tool Works, an invented Sankhya business whose sales track the economy closely. In year 5, the peak, Sarani Tool Works sells Rs 4,00,00,000/-. Costs that move with sales take sixty paise in every rupee of that, or Rs 2,40,00,000/-. Costs that do not move are Rs 1,20,00,000/-. Operating profit is Rs 40,00,000/-. The contribution marginSales less only the costs that move with sales. The remainder covers the costs that do not move, and whatever survives is profit. is Rs 1,60,00,000/-, which is four times the profit, and that ratio of four is the amplifier.

Year 6 arrives and Sankhya output falls 1.00 per cent. Sarani Tool Works sells Rs 3,96,00,000/-, its moving costs fall to Rs 2,37,60,000/-, its unmoving costs stay at exactly Rs 1,20,00,000/-, and profit lands at Rs 38,40,000/-. Profit is Rs 1,60,000/- less, or minus 4.00 per cent. The economy moved one per cent and the profit moved four. Nobody decided that and nobody predicted it. The cost base is doing subtraction, and it would have done the same subtraction in an empty room with the lights off.

What happens to a fixed rupee obligation?

Now change the underlying thing and watch the whole route change with it. A fixed rupee obligation is a promise to hand over a stated number of rupees on stated dates, and that stated number is a nominalStated in rupees of the day, without any correction for what prices have done. The opposite of a figure expressed in what it will actually buy. amount. Nothing in that promise reads the output figure. Sankhya can grow six and a half per cent or shrink two and a half, and the same rupees arrive on the same day either way. So output has no route in.

The price level does have a route in, and it gets in through the back. The rupees are fixed but what a rupee buys is not, so when prices rise the same rupees command less. Take Rs 10,00,000/- promised at the end of Sankhya year 6, a year in which prices rose 8.00 per cent. Dividing Rs 10,00,000/- by 1.08 gives Rs 9,25,926/- of year 5 buying power, rounded to the nearest rupee. The obligation lost Rs 74,074/- of purchasing powerWhat a stated number of rupees will actually buy. Rises when prices fall and falls when prices rise. Converting between the two is taught where inflation is covered., which is 7.41 per cent, rounded to two decimals. Notice the number is not 8.00. The rise sits in the denominator and the fall sits in the numerator, so an 8.00 per cent rise in prices is a 7.41 per cent fall in what a fixed sum buys. Getting that backwards is the single commonest slip in this arithmetic, and it is worked properly where inflation and purchasing power are taught.

Because the rupees are fixed and only the price level can change what they are worth, a fixed rupee obligation's cycle behaviour is mostly an inflation story rather than an output story, and that is exactly why it can move opposite to a profit claim in the very same phase. The two holdings are not reading the same instrument panel. One is watching output through profits. The other is watching prices through what a rupee buys. When output and prices move together the two routes can happen to agree. When output and prices separate, the two routes separate with them, and neither holding has done anything unusual.

What connects a holding with no cash flow at all to the cycle?

None does. A holding that produces no cash flow has an empty middle box. There is no profit for output to reach and no stream of rupees for prices to erode. Its price on any day is set entirely by what a buyer that day is prepared to hand over. A buyer's willingness is a fact about people, not a mechanism running from an output figure.

A story could be constructed anyway. People feel poorer in a contraction, so perhaps they pay less. People fear rising prices, so perhaps they pay more. Both stories are plausible, they point in opposite directions, and neither is anchored to anything that can be computed the way Rs 40,00,000/- to Rs 38,40,000/- was computed above. An invented mechanism gives a reader false confidence in exactly the situation where confidence is least warranted, so saying that no mechanism connects the holding is more useful than inventing one.

The household version is a gold bangle in a locker, or a plot of land nobody farms, or a painting. The bangle pays its owner nothing this year. The effect of a bad year on that nothing cannot be stated without talking about what other people might feel, and feelings are a different kind of statement from arithmetic. An account that pretended otherwise would be dressing up a guess in the clothes of the arithmetic in the sections above, and the arithmetic is the only thing giving those sections their authority.

How can one year hurt two holdings for two different reasons?

Output and prices point opposite ways in Sankhya year 6, so year 6 is the year to look at. Output falls 1.00 per cent while prices rise 8.00 per cent, the fastest in the whole path. The combination has a name, taught where stagflationOutput falling or barely growing while prices rise quickly. Named and worked through where weak growth and high inflation together are taught. is covered. Both routes run through the same twelve months.

Down the profit route, Sarani Tool Works loses Rs 1,60,000/- of profit, from Rs 40,00,000/- to Rs 38,40,000/-, a fall of 4.00 per cent driven by a 1.00 per cent fall in output. Down the fixed rupee route, the Rs 10,00,000/- obligation still pays Rs 10,00,000/-, but buys Rs 9,25,926/- worth of year 5 goods, a fall of 7.41 per cent. Both routes carry damage, and the damage arrives for two entirely unrelated reasons. One holding was hurt by output. The other was not touched by output at all and was hurt by prices. Calling year 6 bad for both without saying why for each is how a reader ends up believing a single dial controls everything.

Now run year 8, the recovery. Output rises 5.00 per cent and prices rise only 3.00 per cent. Sarani Tool Works comes into year 8 off two contraction years, selling Rs 3,86,10,000/- with profit down to Rs 34,44,000/-. Five per cent more sales takes revenue to Rs 4,05,40,500/-, moving costs to Rs 2,43,24,300/-, unmoving costs unchanged at Rs 1,20,00,000/-, and profit to Rs 42,16,200/-. Profit rises Rs 7,72,200/-, or 22.42 per cent, rounded to two decimals. Profit had been squeezed to a smaller base and the same unmoving cost block now looms larger over it, so the amplifier is bigger than it was in year 5, at 4.48 rather than 4.00. Meanwhile Rs 10,00,000/- divided by 1.03 is Rs 9,70,874/-, so the fixed obligation loses 2.91 per cent of what it buys. In the same year the profit route points up 22.42 per cent and the fixed rupee route points down 2.91 per cent. Two holdings move opposite ways with no disagreement between them at all.

Read the two totals rows against each other and then stop. The next step is the one that must not be taken. Neither figure in that table says what any price did. The profit line says what happened inside a set of invented accounts. The buying power line says what happened to an invented sum of rupees. The price somebody would actually pay for a claim on either of them is a different quantity, formed by people who were already looking at the same conditions, and none of the arithmetic above touches it.

Why is every one of these a tendency rather than a rule?

Everything above describes one influence among many. Sarani Tool Works could lose an anchor customer in year 8 and go backwards while Sankhya recovers, or win a large order in year 6 and grow through a contraction. A fixed obligation could be renegotiated, or the promise could fail. The mechanism was never claiming to be the only thing happening. The mechanism claims to be one identifiable route with traceable arithmetic, and it stays true even in a year when something bigger runs it over.

The second reason is sharper and catches careful readers. Whatever anybody would pay for a claim already reflects what people expect. The conditions of year 6 were not a secret while year 6 was happening. Everyone looking at Sankhya could see output falling and prices rising, everyone could do the same subtraction on the same unmoving cost block, and everyone who wanted to act on it had already acted. A mechanism explaining a direction of influence is not a prediction of a movement, and a reader who takes it as one has been misled rather than taught.

Take the everyday version. A vegetable seller knows the wedding season starts in three weeks and demand for a particular flower will jump. So does every other seller in the market, and so does everyone who buys from them. By the time the season arrives, the price already carries the season. A buyer arrives on the first day of the season and says prices should rise now that demand is rising. She has the mechanism completely right and will still be surprised when nothing happens. The rise happened three weeks ago, during the very weeks she spent working out the mechanism. Being right about the direction of an influence and being right about a movement from here are two different achievements, and a mechanism offers only the first.

What does an analyst actually do with all of this?

An analyst covering Sarani Tool Works does something narrower than it looks, and the discipline is worth copying. She takes a phase and refuses to let it stay a phase. Output down 1.00 per cent is not usable until it becomes a line naming customers, an order book, a quarter, and how much of last year's Rs 3,96,00,000/- is actually exposed. Prices up 8.00 per cent is not usable until it becomes a line either: which input costs, on which contracts, repriceable when. If she cannot land an item of macro news on one named row of one named set of books, she files it as background and moves on. Filing as background is most of what she does with most of the news.

A lender doing the same work asks a different question of the same numbers. He is not looking at the profit level, he is looking at whether the unmoving cost block plus the loan repayment can still be covered when the moving costs have already shrunk with sales. In year 6 Sarani Tool Works still cleared its Rs 1,20,00,000/- of unmoving costs with Rs 38,40,000/- to spare. In year 7, with sales at Rs 3,86,10,000/-, that cushion had fallen to Rs 34,44,000/-. The lender watches that cushion thinning, not because a phase told him to, but because the cushion is what his repayment sits behind. Both of them convert a phase into a line item before they let it change anything, and a phase that never reaches a line item never gets acted on.

Neither of them concludes anything about what to hold. The analyst produces an understanding of how a business behaves under conditions; the lender produces a view on whether a specific repayment is safer or less safe. Both stop short of turning that into a position, and they stop there for the same reason: it is a different job requiring different inputs.

What can these mechanisms not establish?

Three things stay out of reach, and each is a genuine boundary rather than a modesty formula.

No mechanism above says what will happen. Every mechanism above runs from a condition that is already known to a consequence for an underlying quantity. None of them runs forward from today to anything. The Sankhya path was written down in full before the arithmetic started, and that is exactly why the arithmetic was possible. No such path exists for any real year in advance.

No mechanism above says what anyone should hold, in any phase. Not in a contraction, not at a peak, not in a recovery, not for a particular reader in a particular situation. The three groups have not been ranked, no phase has been said to favour anything, and no reading of the mechanisms produces a position. Caution is not what stops the answer. Turning a set of influences into a position requires an entirely separate body of work, taught where portfolio construction and allocation are taught. The question belongs there and is answered properly there.

No mechanism above says whether any current price already reflects any of this. Expectation was the whole point of the section above. Everything worked above describes underlying quantities: profit in a set of invented accounts, buying power in an invented sum of rupees. The price somebody would pay for a claim on either is formed by people who already know the conditions, and none of the arithmetic above computes it.