From here the subject is interest rates across time, and the two halves join, because the yield curve is where the single policy rate (M7.02) and the market's inflation and growth expectations get woven together into the entire term structure of rates that actually discounts cash flows.
What is the yield curve?
What the yield curve is. The yield curve (or term structure of interest rates) is simply a plot of the yield (interest rate) on bonds of the same credit quality against their time to maturity. To keep credit risk constant, you build it from the safest borrower in each currency (US Treasury securities for the dollar curve, Government of India securities (G-secs) for the rupee curve), so that the only thing changing along the curve is the maturity, and the curve therefore isolates the pure price of time and interest-rate risk. A typical set of points: the very short end (overnight, 1-month, 3-month bills), the front end (2-year), the belly (5-year, 7-year), and the long end (10-year, 30-year). The 10-year yield is the single most-watched point on both curves: the global benchmark risk-free rate for long-horizon discounting, the number that anchors mortgage rates in the US and corporate-bond and G-sec-linked pricing in India, and the number that (via interest rates and asset prices) sits closest to the discount rate in a long-horizon DCF.
Advertisement
What determines the shape of the yield curve?
What determines the shape: two forces you must be able to name. Why is the curve shaped the way it is on any given day? Decompose any long yield into exactly two pieces:
long yield = average expected future short rate over the term + term premium
- The expectations component. This is the heart of it, and it comes from a simple no-arbitrage idea (the expectations hypothesis): if you can either buy one 10-year bond, or roll over a sequence of short-term bonds for 10 years, then in rough equilibrium the 10-year yield must approximately equal the average of the short rates the market expects over those 10 years. So the long rate is, to a first approximation, the market's forecast of the average future policy rate. This is the profound and useful part: the yield curve is a forecast. An upward-sloping curve means the market expects short rates to rise (economy strengthening, or inflation/policy tightening ahead); a downward-sloping (inverted) curve means the market expects short rates to fall. And since the central bank cuts rates mainly when it expects to need to rescue a weakening economy, an inverted curve is, in effect, the bond market forecasting rate cuts, i.e., forecasting a downturn (see inverted yield curve).
- The term premium. The expectations piece is not the whole story, because a long bond carries risk a rolled-over short position does not: mainly interest-rate (duration) risk (see bond duration) and inflation uncertainty over a long horizon. Investors demand extra compensation for locking their money up long and bearing that risk: the term premium, the additional yield on the long bond beyond the pure expectations average. The term premium is usually positive (which gives the curve a mild upward tilt even when rates are expected to be flat, the liquidity-preference idea of Hicks: people prefer liquid short assets and must be paid to go long), but it is not constant and can even go negative, as it did through much of the 2010s and early 2020s, when enormous central-bank bond buying (QE, M7.02) and a global glut of demand for safe assets bid long yields down below the expectations average, compressing or inverting the term premium. This matters intensely for the inverted yield curve, because a term premium distorted by QE is the leading reason some argue the inversion signal "means less" now. (The New York Fed publishes an estimated term premium, the ACM model, so you can actually watch this component; it is a dashboard item in M7.07.)
Which other theories explain the curve's shape?
Two older theories round out the picture and each explains a real feature: liquidity-preference theory (Hicks) rationalizes the positive term premium; market-segmentation / preferred-habitat theory (Modigliani–Sutch) notes that different investors have strong maturity preferences (banks and money funds want short paper, pension funds and insurers need long paper to match long liabilities), so supply and demand in each maturity "habitat" can push that segment's yield around independently. Preferred-habitat is why QE works (the central bank buying trillions of long bonds shifts the supply-demand balance in the long habitat and pushes long yields down) and why heavy government issuance at the long end can push long yields up regardless of rate expectations (the "supply" story behind some bear steepenings, see curve steepener vs flattener). Hold the two-force decomposition, expectations + term premium, as your master tool: every question about why the curve is shaped this way or why it just moved is answered by asking which of the two pieces changed.
How is India's yield curve different?
The India nuance to carry throughout. The G-sec curve is built and read the same way, but three structural facts make it behave differently from the Treasury curve and must temper any signal you read from it. First, the SLR (M7.02) forces banks to hold ~18% of deposits in G-secs, a large captive, price-insensitive demand base that holds down and distorts yields, especially at maturities banks favour. Second, the RBI's active liquidity operations (OMOs, repos, the corridor) directly manage the short end and intervene along the curve far more hands-on than the Fed does. Third, the India bond market is less deep and less globally arbitraged (though bond-index inclusion is changing this; M7.06). The consequence: India's curve is normally upward-sloping, its inversions are rarer and far less reliable as recession signals than the US curve's, and you should read the level of the 10-year G-sec (the rupee risk-free rate for discounting Indian cash flows) with more confidence than you read the slope of the India curve as a macro forecast. Use the US curve for the global cycle; use the India curve mainly for the rupee discount rate and for RBI-expectation and domestic-liquidity signals.