Economic cycles: why downturns keep coming back

Downturns recur because the economy is a system with delayed feedback. Decisions about credit, inventories and capacity are taken on the basis of the recent past, while their consequences show up much later; as long as conditions are loose, debt and production plans are built up beyond sustainable demand, and the accumulated excess has to be worked off through contraction. What repeats is neither the calendar nor the depth of the fall, but the mechanism itself: the signal arrives late, the response to it arrives with a further delay, and the system overshoots equilibrium in both directions. That is why a cycle cannot be "abolished" by good policy — it can only be smoothed, and an attempt to smooth it harder usually shifts the excess somewhere else.

What repeats and what does not

What repeats is the sequence of states: an expansion of credit and capacity utilisation, overheating, contraction, recovery. What does not repeat is the duration — the intervals between downturns in history differ so widely that any "average" is useless for planning. Nor does the cause repeat: a property crisis, a commodity shock, a banking panic and external restrictions each set off the contraction by a different route. The practical conclusion follows: a phase of the cycle is a description of what is already happening to credit, inventories and margins, not a forecast of the date of the next turn. Publications that name such a date are selling certainty, not knowledge.

The mechanism of excess: credit, inventories, capacity

Credit is the main amplifier. When the interest rate is low and collateral is rising in value, a borrower looks more reliable than they really are: creditworthiness is judged by the price of the asset, and the price of the asset is pushed up by the credit itself. The loop closes, and the quality of debt deteriorates at precisely the moment when the statistics on it look their best.

Inventories add an excess of their own — the bullwhip effect. A retailer sees demand rising and increases its order with a margin; the distributor adds a buffer of its own; the manufacturer expands capacity to match the sum of those orders. When final demand merely stops accelerating, a wave of order cancellations runs down the chain that is many times larger than the change in demand. The capacity built for the peak remains — and becomes a source of losses that cannot be closed quickly.

How the cycle shows up in financial statements rather than in the news

The news describes a downturn once it has already been given a name. In the financial statements it appears earlier and more quietly: profit is still holding up, while cash flow is diverging from it. The gap opens when revenue has been recognised but not collected, and inventories are growing faster than sales. That is why, late in the cycle, the cash flow statement is more informative than the income statement, and in issuer disclosures it is worth looking at working capital and interest expense rather than at the bottom line.

Valuation multiples are deceptive in this phase for a simple reason: the denominator is taken from the cycle. Peak earnings make a valuation look optically cheap, and 3,77 for a cyclical stock at maximum margins says more about the state of the cycle than about the price. This is the core of our guide to P/E: a low reading does not mean cheapness if the earnings cannot be reproduced.

Why the regulator's decision comes late

The regulator responds to data that describe the past, and it influences demand through credit — with a delay that is longer still. As a result, tightening usually continues when the overheating is already behind us, and easing begins when the contraction is already under way. It is exactly this lag that makes the amplitude grow. For the reader, the wording matters more than the rate change itself: it sets the condition for the next step. More on this in how to read the regulator's signal.

What the cycle does to asset prices

The debt market responds to rate expectations, and the sensitivity depends on maturity: the long end of the curve is repriced more sharply than the short end. The mechanism is explained in our piece on duration, and the state of the curve itself can be seen in the quotes for OFZ and the rest of the debt market.

Dividends are not protection against the cycle but a derivative of it: the payout comes out of cash flow, and in the contraction phase it is cut or cancelled. Declared payouts and dates — {{dividend_calendar|limit=5}}, and the full list is in the dividend calendar.

A separate matter is liquidity. It disappears in a non-linear way: the spread widens exactly when a sale has to be made urgently. Planning an exit on the basis of a calm order book is a typical late-cycle mistake.

What this article leaves out

Neither the length of the current phase nor any threshold values for indicators are named here, and that is deliberate: universal thresholds do not exist, and the platform inserts the specific values from its database at the moment of reading. The specifically Russian features — the dependence of export revenue on commodity prices, and the role of the exchange rate and the budget as channels in their own right — are described here as a mechanism; the brief contains no quantitative estimate of each channel's contribution, and inventing one would be worse than saying so plainly. Current market data are gathered in the sections on stocks, events and news, and the terms are in the glossary.

What to watch instead of forecasting a date

Three observable quantities tell you more than any judgement that "we are late in the cycle": the availability of credit and the cost of funding, the ratio of inventories to sales in company reports, and the divergence between profit and operating cash flow. They do not predict the turn, but they show how tightly stretched the system is — and therefore how painful the unwinding of the excess will be once it begins.