Repo-Funded Sovereign Bond Carry and Its Liquidity Risks
Summary
The document describes a fixed-income carry trade in which a trader borrows cash through repo and uses it to buy government bonds. The basic condition for positive carry is that bond income exceed the financing cost. The example discusses cheap short-term funding alongside longer-duration sovereign bonds, and explains how leverage can magnify the apparent return from the yield spread.
The trade can become less attractive if buying pressure raises bond prices and lowers yields, or if central-bank actions withdraw liquidity or raise funding rates. Either change can compress the spread that supports the position and prompt traders to unwind. The account offers a qualitative explanation of the mechanism and its potential ending, but does not quantify collateral haircuts, rollover risk, bond price sensitivity, transaction costs, or losses from adverse rate moves. A positive yield spread therefore does not by itself establish that the leveraged trade is profitable on a risk-adjusted basis.
Key ideas
- Repo-funded bond carry seeks to earn more from bond income than it pays in financing costs.
- Leverage magnifies returns as well as losses on the bond position.
- Bond buying can raise prices and reduce yields, compressing the carry spread.
- Liquidity withdrawal or higher repo rates can weaken the trade and trigger unwinds.
- Rollover, collateral, and interest-rate risks matter beyond the simple yield comparison.
Tags
Full text
# China carry trade — borrow in the repo market and invest in govies # China carry trade — borrow in the repo market and invest in govies > Excess funds in the banking system had juiced leverage in financial markets by driving China’s overnight repo rate to a record low of 0.59% in December. The cheap short-term financing enabled bond traders to make a killing buying sovereign debt with borrowed cash. Source: China crushes leveraged traders with liquidity withdrawals, Bloomberg, 28 January 2021. It seems that for months Chinese traders profited from borrowing through repos and then investing in govies. For this to be possible, does it mean for example China 3-month or 6-month govies had a higher yield than 3m or 6m repos rates? ## Answer by toing (score 1) https://quant.stackexchange.com/a/60857 You are right that for any carry trade to be profitable, cost of funding or leverage cost using repo has to be lower than the investment yields (in this case yields on sovereign bonds). However, for a more detailed explanation, let's start from basics as to why this happens and how does it usually ends. why? If an economy is struggling, central bank through monetary policies attempt to increase money supply. For example, they reduce Fed Fund rate/over night repo rates and introduce other measures of quantitative easing. As a result, money is available for borrowing at cheaper rates in the financial system. what happens then? With easy money supply at lower cost, financial institutions can now borrow this money and invest. They can invest in equities, infrastructure, real investments as well as bond markets. Idea is to increase investment and stimulate economy thereby resulting in growth. What happened here? Some bond traders made use of cheap money and increased leverage, and invested in purchase of sovereign bonds (longer duration like 5 year or 10yr) that would have often 100s of basis point spread .. (i.e. if repo is 3% then may be 10 year govt bonds were 4%). A trader will borrow at 3% and invest in 4% for as long as it can, thus pocketing 1% (referred to as carry) How does it end? If too many people are purchasing same asset (lets say 10yr bond), then price of bond rises and its interest rate reduces. Thus a 4% bond may become 3.35% and therefore profit will reduce. Alternatively, Central bank can reduce liquidity through open market operations or increase over night rate outrightly. In either case, profit from carry will reduce to the extent that its no more worth time and effort on a risk adjusted basis, and the trade ends eventually.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.