Reading the Cycle: Indicators & Expectations

The problem: the economy has no dashboard light

The last few guides gave you the machinery: the [[business-cycle|business cycle]] of boom and [[recession|recession]], the AD-AS diagram, and the idea of an [[output-gap|output gap]] measuring how far the economy sits from its potential. But all of that assumes you can see where you are. In real life you cannot. The official numbers arrive late and get revised for months; a recession is often only declared long after it began. A forecaster's job is closer to a doctor reading scattered symptoms than a driver glancing at a fuel gauge.

So economists assemble a panel of signals, each imperfect, and triangulate. The trick is that different signals carry different timing. Some twitch before the economy turns, some move in lockstep with it, and some only confirm the turn after the fact. Sorting indicators by that timing is the first real skill of cycle-reading, and it is where we begin.

Three timings: leading, coincident, lagging

[[leading-indicators|Leading indicators]] tend to turn before the broad economy does — they are the early tremors. Classic examples: new building permits (people commit to construction months before the work shows up in output), new orders for machinery and equipment, average weekly hours worked (bosses cut hours before they cut heads), stock prices, consumer-confidence surveys, and the spread between long-term and short-term interest rates. That last one — the so-called yield curve — has an unusually good record: when short rates rise above long rates (an "inverted" curve), a recession has often followed within a year or two.

Coincident indicators move roughly in step with the cycle, so they tell you where you are right now: total employment on payrolls, industrial production, real personal income, and real business sales. When economists ask "is the economy expanding this quarter?", these are the numbers they stare at. Lagging indicators only turn after the economy already has — the unemployment rate (firms keep workers on for a while into a recession, then hold off rehiring well into a recovery), the inflation rate, average business-loan interest rates, and the inventory-to-sales ratio. Lagging indicators feel useless for prediction, but they earn their keep by confirming that a turn was real and not a blip.

TIMING        EXAMPLE INDICATORS                 WHAT IT ANSWERS
----------    -------------------------------   -----------------------
leading       building permits, new orders,     where are we HEADING?
              weekly hours, stock prices,
              yield-curve spread, confidence
coincident    payroll employment, industrial    where are we NOW?
              production, real income & sales
lagging       unemployment rate, inflation,     was the turn REAL?
              avg loan rate, inventory ratio

Expectations and Keynes's animal spirits

Notice that several leading indicators — confidence surveys, stock prices, new orders — are really measures of what people expect. That is no accident. Spending today depends heavily on beliefs about tomorrow. A firm builds a factory only if it expects to sell what the factory makes; a household buys a house only if it expects to keep its job. So expectations are not a footnote to the cycle — they are part of its engine.

John Maynard Keynes gave this its famous name. He argued that big investment decisions cannot rest on cool calculation alone, because the future is genuinely unknowable — no honest spreadsheet can tell you the return on a factory thirty years out. What fills the gap, he said, is [[animal-spirits|animal spirits]]: a spontaneous urge to action, a mood of optimism or dread. When animal spirits run high, firms invest and the economy lifts; when they sour, everyone pulls back at once. This bends the tidy picture of pure [[rational-choice|rational choice]] you met early on the ladder — people are not irrational, but under deep uncertainty they lean on confidence and gut feeling, and those move in waves.

Expectations can also become self-fulfilling, which is why central banks obsess over them. If everyone expects a recession, firms cancel orders and households save more, and that very caution causes the slump they feared. The same loop runs through inflation: if workers and firms expect prices to rise 5% next year, they bake 5% into wage demands and price tags, and the expectation delivers itself. That is why [[inflation-expectations|inflation expectations]] are treated as something to be actively managed — a central bank that lets expectations drift loose has half-lost the fight before it starts.

IS-LM in plain words: spending meets money

To turn all this into a forecast you need a way to connect interest rates, spending, and output in one frame. The classic teaching tool is the [[is-lm-model|IS-LM model]], an interpretation of Keynes drawn on two axes: the interest rate going up the side, total output going across the bottom. It is two curves, and the whole thing is friendlier than its initials suggest.

The IS curve is the spending side ("investment equals saving"). It slopes downward for a plain reason: when interest rates are lower, borrowing to build, buy, and invest is cheaper, so total spending and therefore output are higher. Lower rate, more output — that is the whole IS curve. The LM curve is the money side ("liquidity equals money"). It slopes upward: when output is higher, people do more transactions and want to hold more cash, and with the money supply fixed that extra demand for money pushes the interest rate up. More output, higher rate — that is the whole LM curve. Where the two cross sits the economy: one interest rate and one level of output at which both the spending market and the money market are in balance at once.

Now the diagram does useful work, and it dovetails with the AD-AS picture you already own. A government spending boost or a tax cut shifts the IS curve rightward: at any given interest rate there is now more spending, so output rises (this is one of the things behind the [[multiplier-effect|multiplier effect]] you met earlier). A central bank that expands the money supply shifts LM rightward: more money sloshing around lowers the interest rate, which encourages investment and lifts output. In effect, the IS-LM cross is a magnifying glass on the AD side of AD-AS — it shows the gears (interest rates, money, borrowing) that turn a policy lever into a shift in [[aggregate-demand|aggregate demand]].

How forecasters actually read the economy

Put it together and you can picture a real forecasting desk at work — at a central bank, a treasury, or a bank's research arm. Nobody trusts one number. The drill is roughly the same everywhere.

1. Read the coincident indicators to fix where the economy is now — jobs, output, sales — and estimate the current output gap.
2. Scan the leading indicators and expectation surveys for the direction of the next turn, treating each as a hint, not a verdict.
3. Run the story through a model — AD-AS for the big picture, IS-LM for how rates and spending interact — to check the pieces are consistent.
4. Wait for lagging indicators to confirm a turn was real, then revise the forecast as fresh, often-corrected data lands.

Stay humble about the result. Forecasting records are genuinely poor: the profession is notorious for missing turning points, and the deepest reason is the one Keynes pointed at — animal spirits and expectations can flip faster than any indicator updates. There is even a long-running worry, called [[secular-stagnation|secular stagnation]], that the whole cycle may settle into chronically weak demand that the usual signals were not built to flag. Correlation is not causation: an indicator that led recessions for forty years may have been riding some deeper cause that has now changed. The point of all this machinery is not to predict the future precisely — it cannot — but to be a little less blind than you would be staring at one number, and to argue about the economy in a shared, disciplined language.