The weather outside the company

A share is a claim on profits that have not happened yet, and the Treasury yield is what those future dollars are discounted at. When it rises, every one of those claims is worth less today even if the business has not changed at all. It is also what a company pays to refinance: debt raised at 2% that matures into a 5% market costs more to carry, whatever the company did.

• Very cheapunder 2.0%
• Cheap2.0% to 3.5%
• Around average3.5% to 4.5%
• Expensive4.5% to 5.5%
• Dear5.5%+

Expensive: Above the long-run average. Companies refinancing maturing debt do so at a higher coupon than the one rolling off, and future profits are discounted harder.

What would change it: the Federal Reserve moves the short end directly. The long end is decided by what buyers believe about inflation and about how much the government needs to borrow, which is why a 10-year yield can rise on a day the Fed cuts.

The same market, read across instead of over time. Normally it slopes up, because lending for longer carries more risk of being repaid in dollars that buy less. When it slopes down, the market is saying it expects short rates to be cut, and it has said that before every US recession since 1955 - with a lead time from six months to two years, and at least one false alarm. It is a statement about expectations, not a date.

• Deeply invertedunder -0.5%
• Inverted-0.5% to 0.0%
• Flat0.0% to 1.0%
• Normal1.0% to 2.0%
• Steep2.0%+

Flat: Almost nothing extra for lending ten years rather than two. Banks make their money on the gap between short borrowing and long lending, so a flat curve squeezes them.

What would change it: the short end follows the Federal Reserve, so cuts steepen the curve from the left. The long end steepens on its own when investors want more compensation for inflation or for the volume of government borrowing.

Inflation is why rates are where they are, so this chart explains the one above it. Two measures disagree in public every month: CPI is the headline figure and the one written into contracts and pay talks, and core PCE is the one the Federal Reserve actually targets at 2%. Core PCE leaves out food and energy and reweights the basket as shoppers substitute, which is why it is usually the lower of the two. Knowing which is being quoted is most of understanding any inflation argument.

• Falling pricesunder 0%
• Below target0% to 2%
• At target2% to 3%
• Warm3% to 4%
• Hot4% to 6%
• Very hot6%+

Warm: Above target and visible in a weekly shop. The Fed's public language turns to holding rates higher for longer at readings like this.

What would change it: these are year-over-year figures, so a big month dropping out of the comparison moves the rate without anything new happening. Watch the level of prices as well as the rate of change.

The quietest of the four, and the one that shows up in results with no warning. A US company that sells in euros and reports in dollars sees its revenue fall when the dollar rises, having sold exactly as much. It is charted as the change over twelve months rather than as an index level, because 118.6 means nothing without knowing the base year and a 4% move means something immediately.

• Much weakerunder -8%
• Weaker-8% to -2%
• Roughly flat-2% to 2%
• Stronger2% to 8%
• Much stronger8%+

Roughly flat: Currency is doing little to reported results in either direction this year.

What would change it: mostly rates relative to everywhere else. Money moves to where it is paid best, so a dollar yield above Europe's and Japan's pulls the dollar up. How much any one company feels it depends on where its revenue comes from and whether it is hedged, neither of which this chart knows.

The Federal Reserve is charged with two things: stable prices and maximum employment. When they point the same way the decision is easy, and when they conflict this is usually the one that decides it. Two clocks are shown, because the labour market has two: the unemployment rate, monthly, revised and lagging, and initial claims for unemployment insurance, weekly and almost immediate.

Unemployment rate

• Very tightunder 3.5%
• Healthy3.5% to 4.5%
• Softening4.5% to 5.5%
• Weak5.5% to 7.0%
• Recessionary7.0%+

Healthy: A normal working labour market. Wage growth is steady without being a source of inflation.

Initial claims, weekly average

• Lowunder 250k
• Normal250k to 300k
• Rising300k to 400k
• Recession territory400k+

Low: Very few people losing a job in any given week. Consistent with a tight labour market.

What would change it: the unemployment rate can fall because people found work or because they stopped looking, and the survey counts only those still looking. Read it beside claims, which counts an event rather than an answer.

Every series here is produced by a US government agency - the Treasury, the Federal Reserve Board, the Bureau of Labor Statistics, the Bureau of Economic Analysis and the Department of Labor - and retrieved through FRED at the Federal Reserve Bank of St. Louis. Each chart carries its own source and the date of its last reading, and each is also a breakdown with its full history and a data table. How the figures on this site are put together is set out in our method.