The Rate
Machine

Almost everyone carries the same assumption. Higher interest rates are bad for stocks and bad for the economy. It is not wrong so much as incomplete. It describes one channel of a system that has several, and it leaves out the thing that actually determines the outcome.

An interest rate is a price. Specifically, it is the price of money over time, what you pay to use someone else's money now, or what you are paid to postpone spending your own. Once you see it as a price, the question changes shape. You would not ask whether the price of oil being higher is “good” or “bad” without asking why it moved and who is affected. Rates deserve the same treatment.

That single price influences almost everything an economy does.

• Borrowing. Whether a mortgage, a car loan or a factory expansion clears the hurdle
• Saving. What you give up by spending today instead of holding cash
• Investment. Which corporate projects earn more than the cost of funding them
• Consumption. How much of household income is already committed to debt service
• Risk-taking. How far investors reach for return when safe assets pay little
• Asset valuations. What a future stream of cash is worth today
• Currency flows. Where global capital earns the best risk-adjusted return

The case for tightening even when it hurts

Holding rates too low while inflation stays elevated is not a neutral act. Cheap money in an economy already running hot tends to add demand that the economy cannot supply, which pushes prices up further. Worse, it can convince households and businesses that elevated inflation is simply the new normal, and once people plan around that belief, it becomes much more expensive to reverse.

So the honest framing of a rate hike is not “this is bad.” It is a trade. Accepting slower growth now lowers the probability of a more disorderly adjustment later.

But the trade genuinely cuts both ways, and pretending otherwise is its own error. Policy that is too restrictive, or restrictive for too long, slows hiring, starves viable businesses of credit and can turn a soft landing into a recession. There is no setting that carries no risk. There is only the choice of which risk to run.

Money has a price like anything else. When it is cheap, people use more of it. When it is expensive, they use less. Everything else in this guide is an elaboration of that sentence.

Consider a company weighing a $50 million plant. At a 3% borrowing cost the interest bill is roughly $1.5 million a year, and the project needs to clear only that hurdle to be worth doing. At 8% the bill is about $4 million, and a range of projects that made sense at 3% no longer do. Nothing about the plant changed. The price of the money changed, and that was enough.

The same arithmetic runs through a household. A mortgage payment, a car loan, a credit-card balance. Each is a claim on future income, and the rate determines how large that claim is.

This is also why the level of rates, quoted alone, tells you so little. A 5% policy rate was punishingly tight in 2007 and would have been loose in 1981, when inflation was running far higher. What matters is the rate relative to inflation and to the economy's underlying growth, the idea we formalise in Section 14 as the real interest rate.

The Federal Reserve sets a target range for exactly one interest rate, the federal funds rate, what banks charge each other to borrow reserves overnight. As of 16 September 2026 that range is 3.75–4.00%, and the rate actually transacted sits at 3.88%.

That is the entire direct lever. The Fed does not set any of the following.

• Mortgage rates, where the 30-year fixed is near 6.95%
• The 10-year Treasury yield at 4.94%, set by buyers and sellers
• Corporate bond yields, which add a credit premium on top of Treasuries
• Credit-card rates, which are priced off bank funding costs plus a large risk margin
• Stock prices, in any direct sense at all

Every one of those is influenced by the funds rate, and none is determined by it. What connects them is the monetary transmission mechanism, the chain of decisions by banks, companies and households through which one overnight rate eventually reaches the price of a haircut.

A rate hike today does not lower inflation tomorrow. It does not lower it next month either. The effect arrives slowly, through a sequence of human decisions, and by the time it is clearly visible in the data the decision that caused it is a year or more old.

Follow the sequence. The Fed raises rates. Banks reprice loans and tighten terms. A company reviewing next year's capital budget decides to defer one project. A household shopping for a house recalculates the monthly payment and waits. These decisions take weeks or months to make and further months to show up as reduced spending. Only once spending has actually softened do businesses find they can no longer raise prices at will, and only then does inflation moderate.

This is why central bankers talk about forecasting rather than reacting, and why they are so frequently accused of being late in both directions. They are late by construction. The information arrives after the window to act on it has passed.

Inflation is a sustained rise in the general price level. It emerges when the amount buyers want to spend runs ahead of what the economy can produce. When demand exceeds supply, businesses discover they can raise prices without losing customers, and they do.

Monetary policy attacks exactly one term in that equation. Higher rates reduce borrowing, which reduces spending, which reduces demand, which takes away pricing power. Firms that tried to raise prices find customers walking away, and price increases moderate.

The part that is genuinely awkward

Some inflation does not come from excess demand at all. It comes from the supply side, an oil shock, a broken supply chain, a crop failure, a war, a tariff, a shortage of a critical component. Prices rise because the goods are harder to produce or deliver, not because buyers are exuberant.

The uncomfortable truth at the centre of this section is that interest rates cannot produce oil, semiconductors or wheat. A central bank has no instrument that increases supply. Its only lever works on demand.

Which leaves policymakers in a genuinely unpleasant position. Faced with a supply shock, they can either accept higher inflation and risk it becoming embedded in expectations, or suppress demand hard enough to offset a shortage they did not cause and cannot fix. Neither choice is good. This is the trade-off at the heart of stagflation, which we return to in Section 13.

Tightening is usually reported as an act of restriction, something being taken away. There is a case for it that is worth understanding on its own terms, because it explains what a central bank is actually trying to protect.

A · It protects purchasing power

Inflation is a reduction in what your income and savings can buy. At 3.4% a year, money loses roughly a third of its purchasing power over a decade. For anyone holding cash or receiving a fixed income, persistent inflation is a continuous, compounding loss, and one that falls hardest on households with the least ability to hedge it. Price stability is not an abstraction. It is the protection of what people have already earned.

B · It stops expectations from becoming self-fulfilling

This is the argument central bankers care about most, and it is the least intuitive.

Inflation is partly a belief. If workers expect prices to rise 5% next year, they reasonably ask for 5% more pay. If firms expect their costs to rise 5%, they raise prices 5% pre-emptively. Both act on the expectation, and in doing so they produce the inflation they expected. The belief becomes the reality, and at that point inflation no longer needs an original cause. It sustains itself.

C · It protects credibility

A central bank's most valuable asset is being believed. If markets, employers and households are confident that inflation will return to target, they behave in ways that help deliver it, with moderate wage demands, longer-dated contracts and stable long-term rates. Credibility does part of the work so that policy does not have to.

Lose it and everything gets more expensive. This is why a central bank will sometimes tighten into visible economic weakness. The cost of being seen to tolerate inflation exceeds the cost of the tightening itself.

D · It disciplines capital allocation

When money is nearly free, the hurdle a project must clear approaches zero. Businesses that cannot cover their cost of capital survive by refinancing. Investors, unable to earn a return on safe assets, are pushed into riskier ones. Leverage becomes cheap enough to substitute for a sound business case.

Higher rates reverse the filter. Capital becomes scarce enough to be selective, projects must actually earn their cost of funding, and businesses dependent on perpetual refinancing face a real test. The adjustment is genuinely painful. That is precisely what makes it a filter.

E · It rebuilds capacity to respond to the next crisis

This one is easy to miss and matters enormously. Rate cuts are a central bank's primary tool for fighting a downturn. Cutting requires having something to cut from.

A share makes you a part-owner of a business, entitled to whatever is left after everyone else is paid. A bond makes you a lender. You hand over money, receive fixed interest on a schedule, and get your money back on a stated date. No ownership, no upside beyond the agreed interest, and a much stronger legal claim if things go wrong.

A US Treasury security is a loan to the US federal government. Treasuries matter beyond their own market because they are treated as the benchmark “risk-free” asset, the return available without taking credit risk. Nearly every other asset on earth is priced relative to them. A corporate bond yields Treasuries plus a margin for default risk. Mortgage rates track Treasury yields plus a spread. And the discount rate used to value equities starts from a Treasury yield. Move Treasury yields and you move the reference point for everything.

The relationship that confuses everyone

Bond prices and bond yields move in opposite directions, always and mechanically. This is not a market tendency; it is arithmetic.

The reason is that the coupon is fixed for the life of the bond. If the payment cannot change, the only thing that can adjust to new market conditions is the price you pay for it.

Working it through in words

You hold a Treasury with $1,000 face value and a 3% coupon. It pays you $30 a year. Then market yields rise to 5%, so newly issued bonds of the same maturity pay $50 a year.

Now ask the question a buyer would ask. Why would anyone pay $1,000 for your bond paying $30 when they could pay $1,000 for a new one paying $50?

They would not. Your bond is still perfectly good. It will pay every coupon and return the principal, but at $1,000 it is simply worse value. So the price falls until the total return from buying it matches what the new bond offers. In the example above, roughly $846. A buyer at $846 gets $30 a year plus a $154 gain when it matures at $1,000, and that combination works out to about 5% a year.

Nothing about the bond changed. The market's required return changed, and the price moved to meet it. That is the whole mechanism.

The reverse is equally true. If market yields fall to 1%, your 3% coupon becomes attractive, buyers compete for it, and the price rises above $1,000. Move the slider below 3% and you can watch it happen.

One further point has large practical consequences. Longer-dated bonds move much more. A 30-year bond locks in its coupon for thirty years, so a change in required return applies to three decades of payments rather than two years. This sensitivity is called duration, and it is why the 30-year Treasury can lose more value in a rate selloff than most people expect a government bond to be capable of losing.

Here is a recurring source of confusion. The federal funds rate, the 2-year yield, the 10-year yield and the 30-year yield are not the same thing and do not have to move together. The Fed sets the first. The market sets the rest, and each maturity prices a different question.

Why the bond market often moves before the Fed does

Because markets price expectations, not announcements. If investors become convinced the Fed will cut six months from now, they buy bonds today, and yields fall today. When the cut eventually arrives, it changes little, because the price already reflects it.

The chart below is the clearest illustration in this guide. Watch the 2-year yield against the actual policy rate through the last three turning points.

Plot Treasury yields against their maturities and you get the yield curve. Its shape encodes what the bond market collectively expects from growth, inflation and policy. Learning to read it is the closest thing to reading the market's mind.

Normally the curve slopes upward, because lending for longer pays more, compensating for the additional uncertainty. When it does something else, that is information.

The three shapes, and what each one says

• Normal (upward sloping). Short rates below long rates. The market expects growth to continue and policy to stay roughly where it is or drift higher. The unremarkable case.
• Inverted (downward sloping). Short rates above long rates. Investors accept a lower yield to lock money up for a decade, which only makes sense if they expect rates to be considerably lower later, usually because they expect the economy to weaken enough to force cuts. Historically this has preceded most US recessions, though the lead time has varied from months to more than two years, and the 2022–24 inversion was followed by a slowdown rather than the recession most expected.
• Steep. A wide gap between short and long. Typically seen early in a recovery, after a central bank has cut aggressively while long yields price an eventual return to growth and inflation.

Flattening, steepening, and why the direction alone is not enough

The curve flattens when the gap between short and long yields narrows, and steepens when it widens. But two curves can flatten by the same amount for opposite reasons, so the useful question is which end moved.

Financial markets are not a record of what has happened. They are a continuously updated estimate of what is going to happen. Today's price already contains today's news, plus the market's best guess about tomorrow's.

This has an immediate and counter-intuitive consequence. A widely anticipated event usually moves prices very little when it occurs. The anticipation did the moving. What is left to react to is the difference between what happened and what was expected.

Which is why these are both entirely normal

• The Fed hikes and Treasury yields fall. The hike was fully priced, but the statement hinted the cycle is near its end, so the expected path came down even as the current rate went up.
• The Fed cuts and Treasury yields rise. The cut was priced, but officials signalled fewer cuts ahead than markets assumed, or the projections revised inflation upward.

Neither is a market malfunction. In both cases the level moved one way and the expected path moved the other, and the path is what prices assets.

A share is a claim on a company's future profits. To decide what that claim is worth today, you have to convert future money into present money, and future money is worth less, because you could have earned a return on it in the meantime.

$100 in your hand today is worth $100. $100 arriving in ten years is worth less than that. How much less depends on the return you could earn elsewhere, which is called the discount rate. And the discount rate starts from the Treasury yield.

So when Treasury yields rise, the discount rate rises, and the present value of every future dollar of profit falls. Nothing about the business has changed. The arithmetic used to value it has.

Why this hits some companies far harder than others

Look again at the chart in that widget. Raising the discount rate barely touches the first bar and guts the last one. Now apply that to real businesses.

A mature, cash-generative company (a utility, a consumer staples business, a large bank) earns most of its value from profits arriving soon. Those are the early bars. They barely move.

A high-growth company expected to earn very little now and a great deal in a decade holds most of its value in the late bars. Those are the ones that collapse. This is what people mean by long-duration equities, the same concept as bond duration, applied to shares. It is why unprofitable technology companies, early-stage biotech and speculative growth names tend to fall hardest when yields rise, and rally hardest when yields fall, often without any change in their actual business.

But valuation is only one of the two variables

Here is where the reflex “rates up, stocks down” breaks. A share price reflects the profits and the multiple applied to those profits. Higher rates compress the multiple. They say nothing directly about the profits, and if the Fed is hiking because the economy is strong, profits may well be growing at the same time.

“The Fed is cutting” is treated almost universally as good news. Sometimes it is. But a central bank cuts rates for two very different reasons, and the two have close to opposite implications for what you own.

How to tell the two apart in real time

• Speed and size. Gradual 25bp moves at scheduled meetings suggest normalisation. 50bp or larger moves, or anything between meetings, suggests something is wrong.
• Credit spreads. If spreads are tight and stable, the economy is fine. If they are widening quickly, the cut is a response to stress. Section 16 covers this.
• Jobless claims. Rising claims alongside cuts points to the emergency case. Currently they are near 196,000 a week, which is historically low.
• The language. Central banks are careful writers. “Recalibrating” and “moving toward neutral” mean something different from “acting to support economic activity.”

Apply the identical question to tightening. Why is the Fed hiking?

If the answer is that growth is solid, employment resilient, demand strong and inflation somewhat too high, then higher rates are partly a reflection of economic strength. The economy can generate inflation because it is running hot, and the central bank is leaning against it. That is an environment in which corporate earnings can keep growing even as multiples compress, the first box in Figure 16.

This describes the United States right now with reasonable accuracy. The Fed hiked in September 2026 into an economy with unemployment at 4.1%, jobless claims near 196,000, retail sales up 6% year over year, credit spreads at historically tight levels and the S&P 500 near record highs. This is not a central bank fighting a collapse. It is one leaning against an economy with more momentum than its inflation target can comfortably accommodate.

The contrasting case is far less comfortable, because inflation stays high and growth deteriorates.

Stagflation

Stagflation is the combination of stagnant growth and persistent inflation. It is the hardest environment for a central bank because its two objectives point in opposite directions. Cut to support growth and you risk entrenching inflation. Hold or hike to fight inflation and you deepen the downturn. There is no policy setting that addresses both.

It usually originates on the supply side, an energy shock being the classic trigger, as in the 1970s. Supply shocks raise prices and reduce output simultaneously, which is precisely the combination monetary policy is worst equipped to handle. Rates can suppress demand. They cannot create supply.

This is worth watching in the current environment, given oil near $107 and a headline inflation rate well above core. What argues against it for now is the growth side. At 1.5% annualised in the second quarter and 2.1% year over year, growth is slower than it was but not stagnant, and the labour market is firm. The inflation half of stagflation is present. The stagnation half is not.

Every rate quoted in the news is a nominal rate, the headline percentage. The real rate is what remains after inflation, and it is the one that determines whether policy is actually restrictive.

Two examples make the point.

• Nominal 4%, inflation 3% → real rate +1%. Lending money genuinely gains you purchasing power. Borrowing genuinely costs you some.
• Nominal 5%, inflation 6% → real rate −1%. Despite the higher headline number, savers are losing purchasing power and borrowers are repaying in money worth less than they borrowed. Policy is loose, at a 5% nominal rate.

This is why nominal comparisons across time are close to meaningless. A 5% policy rate in 1981, when inflation ran above 10%, was deeply accommodative. A 5% policy rate in 2007, with inflation near 2%, was restrictive. Same number, opposite stance.

The policy rate is a single input into something broader, financial conditions, meaning how easy or hard it is, overall, to obtain and service money in the economy. Several things contribute, and they can pull in opposite directions.

This creates a genuine feedback loop that frustrates central bankers. Suppose the Fed signals it is nearly finished hiking. Equities rally, credit spreads tighten, bond yields fall, and financial conditions ease materially, stimulating the economy the Fed was trying to cool. The market's anticipation of easier policy undoes some of the tightening already delivered. Officials then have to talk tougher to reverse it.

A company borrowing money must pay more than the US government, because it might not repay. That extra yield is the credit spread, and it is the market's live price of corporate risk.

Spreads are a confidence gauge. When investors are relaxed about the economy they accept small compensation for lending to weaker companies, and spreads compress. When they worry about defaults they demand much more, and spreads widen, often sharply, and often before equity markets have fully reacted.

The reason bond investors sometimes notice first is structural. A bondholder's upside is capped at getting repaid, so their entire job is thinking about what could go wrong. Equity investors are paid for what could go right. Given the same ambiguous news, the credit market tends to reprice risk sooner.

The textbook chain is simple enough. Higher US rates make dollar assets more attractive relative to alternatives, global capital moves toward them, and demand for dollars rises.

Interest rates and liquidity are related but not the same. The rate is the price of money. Liquidity is roughly the quantity circulating and how easily it moves. A central bank influences both, using different tools, and they do not always point the same way.

There is a further consequence that matters for the long end of the curve. When a central bank stops buying, someone else must. Government bonds still have to be issued to fund deficits, and private investors will only absorb them at a price they find acceptable. If the amount of issuance is large and the buyer base is price-sensitive, the required yield rises, particularly at longer maturities, where the 30-year currently sits at 5.29%, well above the 3.88% policy rate.

Everything so far has treated the Treasury market as the anchor, the risk-free reference from which other assets are priced. This section examines what happens when investors start questioning that anchor, because it explains a set of market moves that otherwise look inexplicable.

The mechanism

Recall Section 7. When yields rise, existing bond prices fall. Now extend that over time. If long yields rise and then stay high, holders of long-dated Treasuries take a capital loss and then continue holding an asset whose price is not recovering. Add inflation running above target and the real return can be poor for years.

At that point the asset is still risk-free in the narrow sense (you will be repaid), but it is no longer doing the job investors actually want from a safe asset, which is preserving purchasing power. And capital that wants to preserve purchasing power has to go somewhere.

What the data actually shows

This is where it pays to be precise, because the popular version of this argument bundles together assets that have behaved very differently.

Reading that table honestly

The metals case is strong. Gold has risen substantially while real yields sat near two-decade highs, which is the textbook relationship inverting rather than merely weakening. Silver has done the same with more force, and its 0.81 correlation to gold confirms it is trading as monetary metal rather than as an industrial commodity. Both are consistent with capital seeking a store of value outside government paper.

The crypto case is genuinely open. There is real evidence for it. Bitcoin's correlation to gold has moved from roughly −0.10 in the first year of this period to about +0.20 in the most recent one. It is being traded, at the margin, as part of the same complex. That is a meaningful change and worth taking seriously.

But a haven is defined by behaviour under stress, and on that test the evidence is not yet there. Over the past twelve months, precisely the period when long yields rose and the store-of-value question became pressing, gold rose 16% and bitcoin fell 28%. It drew down like a risk asset while the established haven rallied. At 45% annualised volatility it is roughly two and a half times as volatile as gold.

The most defensible reading is that bitcoin is currently trading as a high-beta expression of the same idea rather than as a haven in its own right, closer to how silver behaves relative to gold, but with weaker correlation and larger drawdowns. That is not nothing. It is simply a different claim from “crypto is now a safe haven,” and the distinction matters a great deal for how much of it you would want to own and for what purpose.

The economy is a circuit, not a line. Growth generates employment, employment generates income, income generates demand, demand generates inflation pressure, inflation pressure generates a policy response, and the policy response feeds back into growth. Each stage is caused by the one before it and causes the one after.

Why reality does not follow the circle

The diagram above is a teaching device and should be held loosely. Real economies are pushed around by forces that sit entirely outside the loop.

• Fiscal policy. Government spending adds demand regardless of the policy rate, and can directly offset monetary tightening.
• Deficits and debt issuance. Heavy borrowing raises long yields through supply, independent of the Fed, which is the mechanism behind Sections 18 and 19.
• Productivity. If output per worker rises quickly, the economy can grow faster without inflation. This changes where the neutral rate sits.
• Demographics. An ageing workforce constrains labour supply, which can keep wage pressure elevated through a slowdown.
• Energy prices. A supply shock raises inflation and lowers growth at once, breaking the loop's normal correlation. Oil near $107 is doing this right now.
• Geopolitics and trade policy. Tariffs, sanctions and conflict alter prices and supply chains directly.
• Technology. Large productivity shifts can suppress prices for years and change what a “normal” rate looks like.
• The banking system. If banks are impaired, transmission breaks down entirely, and rate changes stop reaching the real economy.

Which is why cycles vary so much in length and character, why every recession has a different trigger, and why confident macro forecasts have such a poor record. The loop is a map of the mechanism. It is not a schedule.

Almost any macro environment can be located using two questions. Is growth strong or weak? and is inflation high or low? Crossing them gives four quadrants, each with its own characteristic policy response and market behaviour.

This is the single most portable tool in the guide. It will still work in twenty years.

Two cautions on using it. First, the boundaries are fuzzy. Economies drift between quadrants rather than jumping, and the current US reading is genuinely mixed, with growth solid but decelerating and core CPI is near target while core PCE is not. Second, the behaviours listed are tendencies drawn from history, not rules. Markets have surprised in every one of these quadrants.

Seven categories cover the macro picture. Each cell below shows the current reading and, more importantly, the question you are asking it, because the question is what remains constant when the numbers change.

Sources: BLS, BEA, Federal Reserve, US Treasury, ICE BofA and EIA via FRED; equity, gold, silver and crypto prices from Yahoo Finance. Latest available observation for each series as of 20 September 2026; reporting dates differ by series.

A single economic release is almost always ambiguous. The same number can be good news or bad news depending entirely on what sits beside it. This is the most common way confident analysis goes wrong.

Example one · inflation falls

Encouraging? It depends completely on the cause.

To tell them apart, look at what accompanies the inflation print, meaning retail sales, employment and industrial output. Falling inflation alongside a firm labour market is the benign case. Falling inflation alongside rising jobless claims is not.

Example two · unemployment rises

A rising unemployment rate helps bring inflation down through less wage pressure, less spending and less pricing power. A central bank fighting inflation may quietly regard it as progress.

But the rate of change matters more than the level. Unemployment drifting from 4.1% to 4.3% over a year is a labour market coming into balance. Moving from 4.1% to 5.0% in six months is something else entirely, because job losses feed on themselves. Unemployed people spend less, which reduces revenue, which causes more layoffs. Labour markets rarely deteriorate gently once they begin.

Example three · the current US data, read properly

The live illustration is in this guide's own numbers. Core CPI is 2.45%, which looks like victory. Core PCE is 3.34%, which does not. Headline CPI is 3.35%, but oil is up 68%, so much of that is supply. Meanwhile breakevens sit at 2.33%, saying the market expects all of it to resolve.

No single one of those numbers describes the situation. Read together, they say something fairly precise. There is a real energy shock passing through headline inflation, underlying inflation is somewhat above target on the measure the Fed actually uses, and expectations remain anchored. That combination justifies caution without justifying alarm, which is a reasonable description of what the Fed did.

Every macro event sets off a chain. The first link is obvious and gets priced within minutes. The links further down take longer to appear and are where most of the analytical value sits.

This is the procedure. It works for an FOMC decision, a CPI print, a payrolls report or anything else that moves markets, and it does not expire.

Why the surprise matters more than the number

A CPI print of 3.2% is bullish if the market expected 3.5% and bearish if it expected 2.9%. The absolute figure is nearly irrelevant to the immediate market reaction. What moves prices is the gap.

This is why markets sometimes rally on objectively poor data. The data was poor, but less poor than feared, and the fear was already in the price.

On 16 September 2026 the FOMC raised its target range by 25 basis points to 3.75–4.00% on a unanimous 12–0 vote, the first increase in more than three years. The statement said inflation “remains elevated” and that the move would “support a timelier return to the Committee's 2 percent goal.” The accompanying projections indicated most officials expect the rate to end 2026 between 4.1% and 4.4%, implying at least one further hike.

What happened to get here

The Fed cut through late 2025, taking the effective rate from 4.08% in September to 3.63% by December, then held for nine months. Over that hold, inflation re-accelerated. Headline CPI rose from 2.39% in January to a peak of 4.17% in May, and core PCE, the Fed's actual target, climbed from 2.75% in October 2025 to 3.34% by July 2026. The September hike is a response to that re-acceleration.

The case that tightening is currently justified

• Inflation is above target on the measure that counts. Core PCE at 3.34% is not close to 2%, and it has been rising, not falling, for most of a year. Core CPI at 2.45% looks better, but the Fed does not target it.
• The economy can absorb it. Unemployment is 4.1% and falling, claims are near 196,000, retail sales are up 6% year over year. This is the box-one case from Section 13, tightening into strength rather than into weakness.
• Financial conditions are not restrictive. High yield spreads at 2.70%, investment grade at 0.78%, banks neutral on lending standards, equities near record highs. Policy is tight; conditions are not. That gap is an argument for more tightening, as covered in Section 15.
• Expectations are still anchored, which is the moment to act. The 10-year breakeven is 2.33%, slightly below where it stood a year ago. Defending an anchor that is holding is far cheaper than restoring one that has slipped. This is Section 6B in practice.
• It preserves room for the next downturn. As Section 6E put it, a policy rate near 4% is four percentage points of conventional ammunition. A rate near zero is none.

The case for caution, where this could go wrong

• Much of the inflation is supply-driven, and rates cannot fix supply. Oil at $107, up 68% year over year, explains a large share of the gap between headline and core. Suppressing demand to offset an energy shock destroys activity that was never the problem. This is Section 5's central difficulty.
• The labour market is firmer at the surface than underneath. Unemployment at 4.1% looks robust, but payroll growth has averaged roughly 71,000 a month over the last three months, well below the pace of early 2026, and monthly prints have swung between −156k and +214k this year. Hiring is decelerating even as the unemployment rate falls.
• Growth is already slowing. Real GDP grew 1.5% annualised in Q2, down from 2.1% in Q1. The Fed is tightening into a decelerating economy, and Section 4's lags mean the full effect of this hike arrives in late 2027.
• The long end is already doing work. The 30-year at 5.29% and the 10-year real yield at 2.61% are genuinely restrictive levels, and the 30-year mortgage at 6.95% is pressing directly on housing. Rate increases at the front end compound restriction that already exists further out.
• The market may be pricing more than the economy can take. The 2-year at 4.67% implies roughly 79bp of additional tightening. If the Fed delivers that and growth is weaker than it appears, the over-tightening risk from Section 1 becomes real.

The honest summary

This is a genuinely balanced situation, and anyone presenting it as obvious in either direction is skipping evidence. The Fed is tightening into an economy that is still growing, with employment intact, expectations anchored and no credit stress, conditions that make tightening defensible and survivable. At the same time it is doing so against inflation that is substantially energy-driven, with growth already slowing, hiring decelerating and long-end rates already restrictive.

The tension identified in Section 22's dashboard is the thing to hold onto. The Fed is behaving as though this economy needs restraining, and the credit market is behaving as though nothing is wrong at all. Those two positions cannot both be right indefinitely. Which one resolves, and how, is the question that matters over the next several quarters.

Rather than forecasting, define the branches in advance and attach observable triggers to each. Then let the data tell you which one you are on. This is the same two-variable logic as Section 21, applied forward.

Whenever interest rates make a significant move, in either direction, for any reason, work through these in order. The current readings are shown so you can see the method applied, but the questions are the durable part.

If the guide has worked, one reflex has changed. You no longer hear “the Fed raised rates” as a verdict. You hear it as the visible output of a system, and you know which questions unlock the rest.