India’s long-term bond yields less sensitive to inflation in AI era: NIPFP

The NIPFP study found a clear change after September 2022. Long-term government bond yields became less responsive to expected inflation. Short-term rates became more responsive. This matters because yields influence financial conditions and the cost of government borrowing.

Researchers identified September 2022 as a structural break. They examined 10-year, five-year and three-year government securities, along with 91-day Treasury bills. The study also tested the relationship between yields, expected inflation and broad money supply, or M3. Better information and forecasting may have reduced uncertainty at the long end.

The change cannot be credited to AI alone. Post-pandemic policy normalisation, a global inflation surge and geopolitical tensions occurred during the same period. The government’s borrowing costs will still depend on fiscal supply, investor demand, RBI policy and the term premium, even if inflation becomes less influential at the long end.

A government bond yield is the return associated with holding a government security. It also reflects the government’s borrowing cost in the market. A higher yield generally means investors demand more compensation for lending. Yields matter because they influence financial conditions and the expense of government borrowing.

Short-term yields, such as those on 91-day Treasury bills, are closely linked to current monetary policy and liquidity conditions. Long-term yields, such as those on 10-year securities, also reflect expected inflation, future interest rates, investor demand and the term premium. The term premium is the extra return investors require for locking money away longer.

That is why the two ends can move differently. The NIPFP study found long-term yields became less responsive to inflation expectations after September 2022, while short-term rates became more responsive. The article cautions that borrowing costs still depend on several market forces, not inflation alone.

The study used monthly observations covering 2000 to 2025. That long span allowed researchers to compare bond-market behaviour across different economic conditions, including the period after the pandemic. They were looking for changes in how yields related to expected inflation and broad money supply, or M3.

The analysis covered four maturity categories. It examined 10-year, five-year and three-year government securities, which represent longer and medium-term borrowing periods. It also included 91-day Treasury bills, a short-term government instrument. Together, these securities helped researchers compare the short and long ends of the sovereign yield curve.

The researchers used an autoregressive distributed lag model. They identified September 2022 as a structural break and tested whether relationships changed after that point. The study found different responses at the two ends, but it did not prove that AI alone produced the change.

A yield curve compares the yields investors demand on government securities with different maturities. It shows how borrowing costs vary from very short periods to several years. The curve matters because it helps reveal how markets view inflation, monetary policy, liquidity and risks over time.

For example, the NIPFP study compared 91-day Treasury bills with three-year, five-year and 10-year government securities. It found that after September 2022, short-term rates became more responsive to inflation expectations, while long-term yields became less responsive. That difference can widen the gap between the two ends of the curve.

A widening gap deserves careful attention, according to Madhavi Arora. It might indicate changes in inflation expectations, the way monetary policy reaches markets or the term premium. It does not automatically show that borrowing costs have fallen. Fiscal supply and investor demand remain important influences on long-term yields.

AI tools can help regulators and policymakers detect risks earlier and forecast banking-system liquidity more accurately. If markets receive better information, investors may feel less uncertain about future financial conditions. That could weaken the effect of inflation expectations and liquidity conditions on long-term bond yields.

The proposed mechanism is straightforward. Improved risk detection may help authorities anticipate problems, while better liquidity forecasts may make funding conditions easier to assess. Investors could then demand less additional compensation for uncertainty linked to those factors. In the NIPFP study, this possibility helps explain why long-term yields became less responsive after September 2022.

However, the researchers did not establish that AI caused the change. The same period included policy normalisation after the pandemic, a global inflation surge and geopolitical tensions. Long-term yields still depend on government borrowing requirements, investor demand, RBI policy and the term premium.

The study cannot isolate AI as the sole cause because September 2022 marked more than a possible change in technology use. It also came during a period when economies were moving away from pandemic-era policies. Monetary policy was being normalised, which could directly change short-term rates and broader financial conditions.

At the same time, global inflation surged and geopolitical tensions increased. These forces can influence inflation expectations, liquidity conditions, investor demand and the compensation investors require for holding longer-term bonds. Any of them could have affected the relationship between yields and the variables examined in the study.

The researchers therefore describe AI as a possible contributor, not a proven explanation. Their model identifies a structural break and a changed relationship after September 2022, but it does not establish causation. The article also says yields continue to depend on RBI policy, government borrowing needs, investor demand and the term premium.

When long-term government bond yields remain high, the Union government generally faces a higher cost when raising money through market borrowings. That can make its borrowing programme more expensive. A weaker link between long-term yields and inflation expectations therefore does not guarantee cheaper borrowing.

Inflation can increase the return investors demand, while the RBI’s policy stance can influence financial conditions and liquidity. Heavy government borrowing can add supply to the bond market. If investor demand is weak, yields may need to rise to attract buyers. The term premium adds compensation for locking money away for longer and for facing longer-term risks.

The article highlights this combination as the practical issue for the Centre. Long-term yields may respond less to inflation expectations, but fiscal supply, investor demand and the term premium remain crucial. The study also does not show that AI has lowered borrowing costs. RBI policy and liquidity management remain important for financial conditions.