The Yield Curve and What It Actually Predicts

Yield Curve Theory | Ind AS 109 ECL | Ind AS 19 DBO | Ind AS 116 IBR

One curve. Three theories. And three Ind AS standards whose most critical inputs depend on getting the reading right.

A CFO approves an Ind AS 19 actuarial valuation. The actuary has used a discount rate of 7.1% - derived from the yield on government bonds "of similar duration." Three doors down, the treasury team is using the same government securities curve to price an Ind AS 109 ECL model. Across the corridor, the finance team is computing an Ind AS 116 incremental borrowing rate for a new lease. All three are reading the same curve. All three are reading it differently. One of them is wrong. 

The yield curve - the graphical relationship between the yield to maturity of fixed-income instruments and their time to maturity - is one of the most information-dense objects in finance. It is simultaneously a snapshot of current interest rate conditions, a market forecast of future rates, a measure of credit risk across tenors, and a reflection of central bank policy. For a finance professional working under Ind AS, the yield curve is not merely a market data point. It is a direct input into at least three accounting standards - and misreading it produces misstated financial statements.

This article maps the three principal theories of the yield curve, explains what each predicts and what it does not, and then applies the framework precisely to the discount rate requirements of Ind AS 109, Ind AS 19, and Ind AS 116.

Part 1: The Three Theories of the Yield Curve

The shape of the yield curve - whether it slopes upward (normal), downward (inverted), flat, or humped - is explained by three competing theories. They are not mutually exclusive; in practice, each captures a different dimension of the curve's behaviour.

Theory 1: Pure Expectations Theory

The Pure Expectations Theory (also called the Unbiased Expectations Theory) holds that the long-term interest rate is simply the geometric average of current and expected future short-term rates. A five-year bond yield is not independently priced - it is the market's best estimate of what rolling one-year bonds over five years would yield.

• What the Pure Expectations Theory predicts: An upward-sloping yield curve implies that the market expects short-term rates to rise in the future. A flat curve implies no expected change. An inverted curve implies the market expects short-term rates to fall - which historically occurs when the market anticipates a recession and consequent central bank easing. Under this theory, there is no risk premium for holding longer-duration bonds - the entire term structure is driven by rate expectations alone.
• What it does not explain: In practice, investors consistently demand a premium for holding longer-duration bonds - even when rate expectations are flat. The theory cannot explain a persistently upward-sloping curve in a stable rate environment. This gap is addressed by the Liquidity Preference Theory.

Theory 2: Liquidity Preference Theory (Keynes/Hicks)

The Liquidity Preference Theory - formalised by Hicks (1946) - modifies the Pure Expectations Theory by adding a liquidity premium to longer-maturity yields. Investors prefer shorter-duration instruments because they carry less interest rate risk (price sensitivity). To induce investors to hold longer bonds, the market must offer a premium above the pure expectations-implied rate.

• What the Liquidity Preference Theory predicts: The yield curve will almost always slope upward, even when rate expectations are flat, because the liquidity premium is always positive and increasing with maturity. An inverted curve under this theory is particularly significant - it implies that the market's expectation of future rate declines is so strong that it more than offsets the positive liquidity premium. Historically, an inverted curve (after stripping out the liquidity premium) is one of the most reliable leading indicators of recession.
• The Indian Context: The RBI's liquidity operations - repo rate, reverse repo, Standing Deposit Facility, Open Market Operations - directly influence the short end of the G-Sec curve. The term premium on longer G-Secs (10-year, 30-year) reflects both liquidity preference and fiscal supply concerns (the government's borrowing programme). This is why the Indian G-Sec curve can steepen sharply even during rate-hold cycles - not because rate expectations have changed, but because the liquidity premium has widened due to increased government supply at the long end.

Theory 3: Market Segmentation Theory

The Market Segmentation Theory takes the most radical position: different maturity segments of the yield curve are essentially separate markets, driven by the supply and demand of investors with specific maturity preferences. Insurance companies and pension funds prefer long-duration assets (to match long-duration liabilities). Banks prefer short-duration assets. The yields at different tenors are therefore determined independently - the 1-year yield and the 30-year yield can move in different directions simultaneously without implying anything about future rate expectations.

The Preferred Habitat Variant: The Preferred Habitat Theory (Modigliani and Sutch, 1966) softens the segmentation assumption - investors do have preferred maturity habitats, but they can be induced to move to other maturities if the yield premium is sufficient. This is the most empirically supported theory for the Indian market, where LIC, EPF, and insurance companies dominate the long end of the G-Sec curve, while mutual funds and banks dominate the short end.

Part 2: Ind AS 109 - The ECL Discount Rate

Paragraph 5.5.17 of Ind AS 109 requires that expected credit losses be discounted at the effective interest rate (EIR) of the financial asset - or an approximation thereof. The EIR is the rate that exactly discounts the estimated future cash flows of the financial asset back to its gross carrying amount at initial recognition. For a fixed-rate instrument, the EIR is constant. For a floating-rate instrument, the EIR is periodically revised as the benchmark rate changes.

Why the Yield Curve Matters for ECL

The ECL discount rate is not a market rate chosen at the measurement date - it is locked at origination. This creates a fundamental interaction with the yield curve:

01. Fixed-Rate Instruments: The EIR is set at origination and does not change. An instrument originated when the yield curve was steep (high long-term rates) carries a high EIR permanently. As the yield curve flattens or inverts, the discount rate for ECL purposes does not move - the ECL charge is being discounted at a historical rate, not the current market rate. [Ind AS 109, Para 5.5.17; Appendix A (EIR definition)]

02. Floating-Rate Instruments: The EIR is updated each period to reflect changes in the benchmark rate. For a MCLR-linked loan or a SOFR-linked instrument, the EIR in the ECL calculation changes as the benchmark changes. A steep yield curve environment - with high short-term rates - means a higher EIR and therefore a lower present value of credit losses (same cash loss, higher discount rate). As the curve flattens or inverts and short rates fall, the EIR falls and the present value of losses rises. This is a mechanical ECL movement that has nothing to do with credit quality. [Ind AS 109, Para B5.4.5]

03. Trade Receivables (Simplified Approach): For entities using the simplified approach under Paragraph 5.5.15 of Ind AS 109, the discount rate is often ignored in practice - especially where the credit term is short (30–90 days). The standard permits an approximation where the time value of money adjustment is immaterial. However, for longer credit terms (180+ days, which are common in infrastructure and government receivables in India), omitting the discounting is not immaterial and constitutes a departure from the standard. [Ind AS 109, Para 5.5.15; BC5.260]

The yield curve tells the ECL modeller: (a) what the current EIR is for new floating-rate originations, (b) how the existing EIR on floating-rate instruments will change as the benchmark moves, and (c) whether the time value adjustment on long-dated trade receivables is material enough to require explicit discounting. The Pure Expectations Theory is most relevant here - the forward rates implied by the G-Sec curve are the market's forecast of future benchmark rates, which directly drive future EIRs on floating-rate instruments and therefore the path of future ECL charges.

Forward Rates from the Yield Curve - The ECL Connection

For stage 2 and stage 3 assets requiring lifetime ECL, the cash flows being discounted extend over multiple years. The appropriate discount rate for each year's expected cash flow is - technically - the forward rate for that period implied by the yield curve at origination. Using a single flat EIR is an approximation. For material long-dated exposures, the forward rate structure matters.

The Liquidity Preference Theory is critical here: the forward rates implied by a normal yield curve are not unbiased forecasts of future spot rates - they are biased upward by the liquidity premium embedded in the curve. An ECL model that uses raw forward rates from the G-Sec curve as its expected future benchmark rates will systematically overstate the EIR for floating-rate instruments in a stable rate environment. The correct approach is to strip the liquidity premium from the forward rates before using them as rate forecasts - a step that almost no non-bank ECL model performs.

Part 3: Ind AS 19 - The Defined Benefit Obligation Discount Rate

Paragraph 83 of Ind AS 19 states that the rate used to discount post-employment benefit obligations shall be determined by reference to market yields at the end of the reporting period on high-quality corporate bonds. In countries where there is no deep market in such bonds, the market yields on government bonds shall be used. India falls into the second category - the RBI does not recognise a "high-quality corporate bond" market of sufficient depth for this purpose. Therefore, Indian entities use the G-Sec yield as the Ind AS 19 discount rate.

The Tenor Matching Requirement

Paragraph 83 further specifies that the currency and term of the bonds used to determine the discount rate shall be consistent with the currency and estimated term of the post-employment benefit obligations. This is the duration matching requirement - and it is almost never applied correctly in practice.

The defined benefit obligation for a gratuity or provident fund scheme has a duration - a weighted average time to payment - that depends on the age distribution, expected tenure, and salary escalation of the workforce. For a company with a young workforce, the duration of the DBO might be 15–20 years. For a mature workforce nearing retirement, it might be 5–8 years. The discount rate must be the G-Sec yield at a tenor matching this duration - not the 10-year G-Sec yield by default, which is the actuarial shorthand used by most practitioners.

The Market Segmentation Theory is directly relevant here. The 30-year G-Sec yield in India is heavily influenced by the purchasing behaviour of LIC - which dominates demand at the long end of the curve as part of its liability matching strategy. When LIC absorbs large issuances of 30-year paper, yields at that tenor compress below what the Pure Expectations Theory would predict. An actuary using the 30-year yield as the Ind AS 19 discount rate for a young workforce is using a rate that has been suppressed by institutional demand - not a rate that reflects the market's true expectation of long-run interest rates.

The OCI Treatment - And Why the Yield Curve Creates Volatility

Under Paragraph 120 of Ind AS 19 , remeasurements of the defined benefit liability - which include actuarial gains and losses arising from changes in financial assumptions (including the discount rate) - are recognised immediately in Other Comprehensive Income (OCI) and are never reclassified to profit or loss. This is a critical departure from the "corridor approach" of old Indian GAAP.

The consequence: every movement in the G-Sec yield curve at the year end - even by 25 basis points - produces a remeasurement that flows directly to OCI and therefore to the balance sheet equity. For a company with a large DBO, a 50 basis point rise in the G-Sec yield (reducing the DBO) produces a significant actuarial gain in OCI. A 50 basis point fall (increasing the DBO) produces an actuarial loss. The yield curve, in other words, produces balance sheet volatility in the equity of every company with a defined benefit scheme - even if service cost and interest cost in P&L remain stable.

Ind AS 19 requires the discount rate to be the G-Sec yield at the end of the reporting period. For a March 31 year-end, this is the March 31 closing yield - not the average yield for the year, not the yield on the valuation date if the actuary visits in February. In years where the G-Sec yield moves significantly between February (when most actuarial valuations are prepared) and March 31 (the measurement date), the actuarial report must be updated to reflect the closing yield. Many entities use the February yield and do not roll forward - producing a discount rate that does not comply with Paragraph 83 of Ind AS 19.

Part 4: Ind AS 116 - The Incremental Borrowing Rate

Paragraph 26 of Ind AS 116 defines the Incremental Borrowing Rate (IBR) as the rate of interest that a lessee would have to pay to borrow over a similar term, and with a similar security, the funds necessary to obtain an asset of a similar value to the right-of-use asset in a similar economic environment. The IBR is used where the rate implicit in the lease cannot be readily determined - which is the case for almost all operating leases converted to finance leases under Ind AS 116.

The Four Dimensions of the IBR

The IBR is not a market rate - it is a lessee-specific rate that must reflect four simultaneous conditions:

01. Similar Term: The tenor of the IBR must match the lease term (including reasonably certain renewal periods). A five-year lease requires a five-year borrowing rate - not the company's weighted average cost of debt, which blends all maturities. [Ind AS 116, Para 26; Appendix A]

02. Similar Security: The IBR should reflect a secured borrowing rate, because the right-of-use asset effectively serves as collateral for the implicit lease obligation. An unsecured borrowing rate overstates the IBR and therefore understates the lease liability and ROU asset. [Ind AS 116, Para 26; IFRS 16.BC160]

03. Similar Value: The amount borrowed should be comparable to the ROU asset value - a very large lease on a single asset may attract different terms than a small lease. [Ind AS 116, Para 26]

04. Similar Economic Environment: The IBR must be determined in the currency of the lease, in the economic environment of the lessee, at the commencement date of the lease. [Ind AS 116, Para 26]

The yield curve theories bear directly on each component:

• The G-Sec base rate embeds a liquidity premium per the Liquidity Preference Theory. For a 10-year lease, the 10-year G-Sec yield includes a term premium that reflects duration risk, not just expected future rates. Using the Pure Expectations Theory to argue that the "true" base rate is lower (by stripping the term premium) would reduce the IBR - but this is not standard practice and is not supported by Ind AS 116's reference to market borrowing rates.

• The credit spread should reflect the lessee's actual credit quality - not an investment-grade proxy. For an unrated Indian company, this requires reference to comparable rated issuers' borrowing spreads over G-Secs at the same tenor. Market Segmentation Theory is relevant: the corporate bond market at long tenors (7–10 years) is thin in India, and the spread may not reflect a genuinely traded market clearing rate.

• For USD-denominated or foreign currency leases entered by Indian entities, the IBR must be constructed in INR terms by converting the foreign currency rate using the covered interest rate parity relationship - adjusting for the forward premium or discount between INR and the lease currency.

THE COMMON IBR ERROR: USING WACC

The single most common Ind AS 116 IBR error in practice is using the company's Weighted Average Cost of Capital (WACC) as the incremental borrowing rate. WACC blends equity cost and debt cost across all maturities and capital structure layers. It is neither lessee-specific in the required sense, nor tenor-matched, nor does it reflect a secured borrowing rate. It is almost certainly wrong. A company with a WACC of 12% that secures a 5-year lease with G-Secs at 7.0% and a credit spread of 150bps should use an IBR of approximately 8.5% - not 12%. The difference produces a materially higher lease liability and ROU asset at inception.

Part 5: Reading the Same Curve - A Unified Framework

The three Ind AS standards read the same underlying G-Sec yield curve - but they read it differently, at different points, for different purposes. The discipline is understanding which theory applies to each reading:

The yield curve is not a single number. It is a term structure - a function of maturity. Every time an Ind AS standard asks for a 'market rate' or a 'risk-free rate' or a 'borrowing rate,' it is asking for a specific point on a specific curve - constructed in a specific way. Getting the point wrong by even 50 basis points produces a material misstatement

Closing Thought

The yield curve is the most consequential input in financial reporting that most preparers treat as a single number. It is not. It is a term structure with a shape that each of the three theories explains differently - and each explanation has a different implication for how the curve should be read for accounting purposes.

A CA or CFO who understands that Ind AS 109's EIR is a historical rate locked at origination, that Ind AS 19's discount rate must be duration-matched and point-in-time at the reporting date, and that Ind AS 116's IBR must reflect a lessee-specific secured rate at the correct tenor - and who can explain the yield curve theory behind each of these requirements - is not just technically compliant. They understand why the standard is written the way it is. That understanding is what separates a technically rigorous practitioner from one who fills in a rate and moves on.