Bond-CDS basis trade Japan

TL;DR

The Japan bond-CDS basis is the spread between a corporate’s CDS protection cost and its cash bond’s credit spread over the swap or JGB benchmark curve at matched tenor. Negative basis (CDS < bond spread) historically attracted long-cash-short-CDS trades that lock in a “free” carry net of funding cost and balance-sheet charges. Positive basis (CDS > bond spread) is less common in Japan but appears during stress when protection-buying demand exceeds dealer capacity. Post-2010, Basel-III leverage ratio, single-counterparty credit limits, repo specials on JGBs, and BOJ unconventional policy (NIRP, QQE, YCC, YCC adjustments) have made basis trades harder to scale and more volatile through cycles. Major basis dislocations in Japan IG corporate credit occurred during COVID Q1 2020 (negative basis blew wide as cash bonds traded distressed faster than CDS), 2022-2023 BOJ YCC adjustments (rate and spread repricing widened basis), and 2024 episodic Japan financial credit stress (smaller dislocations on selected names).

Wiki route

This entry sits under derivatives index as the arbitrage / relative-value layer that complements Japan CDS market overview and Japan corporate CDS spread mechanics. Read it together with Japan IRS market for the swap-curve anchor used in asset-swap pricing, yen basis swap market for the JPY funding-cost layer, and cross-currency basis swap Japan for the USD-JPY funding interaction that affects USD-funded participants in JPY basis trades.

On the cash-bond side, cross-reference finance index, convertible bond mechanics, and money-market index for the JGB repo and funding context. On the institutional side, banking index, Japan life insurance ALM, and prime brokerage and institutional financing map the dealer / fund counterparties of the trade.

Definition

Bond-CDS basis = CDS spread - Bond Z-spread (or asset swap spread) at matched tenor

Sign	Interpretation
Negative basis	CDS spread < Bond spread → buying protection is cheaper than the bond’s credit spread implies → long-cash-short-CDS attractive
Zero basis	Pure-arbitrage equilibrium (theoretical)
Positive basis	CDS spread > Bond spread → selling protection captures more than the bond’s credit spread → short-cash-long-CDS attractive

Why basis is not zero

In a frictionless world the basis would be zero. Real-world frictions create persistent and time-varying basis:

Friction	Effect on basis
Bond funding cost (repo rate, balance-sheet charges)	Tends to push basis negative (bond costs more to hold than synthetic)
Credit-event coverage difference (deliverable obligation universe, restructuring scope)	CDS coverage may be broader or narrower than bond exposure
Cheapest-to-deliver option on CDS	Adds value to CDS protection vs cash bond, pushing basis positive
Counterparty risk on CDS	Reduces value of CDS protection, pushing basis negative
Liquidity asymmetry	Less-liquid leg trades at premium / discount
Regulatory capital cost	Bank-owned long-cash-short-CDS positions consume risk-weighted assets and leverage capacity
Tax and accounting	Different recognition timing of cash bond income vs CDS premium

Structure

Leg	Action
Long cash bond	Buy reference issuer’s bond at par or near-par, funded in repo
Short CDS protection (i.e., buy protection)	Pay running spread for protection against credit event
Net carry	Bond yield - Funding rate - CDS premium

Provided the negative basis exceeds funding cost and balance-sheet charges, the trade locks in positive carry with credit-event protection from the CDS leg.

P&L decomposition

Daily P&L ≈ (Bond accrual - Repo funding cost) - (CDS premium accrual)
           + Mark-to-market change in (Bond price - CDS protection value)

If the basis narrows toward zero, the position appreciates (bond rises faster than CDS, or CDS widens faster than bond). If basis widens further negative, position loses MTM.

Required holding period

To realize the locked-in carry, the position typically needs to be held to bond maturity or to a known horizon. Short-horizon trades depend on basis convergence which is unpredictable. Many funds historically rolled positions and accepted MTM volatility.

Funding cost considerations

The bond leg requires funding. Funding cost is the dominant determinant of trade economics:

Funding source	Cost driver
Repo (bilateral or tri-party)	Repo rate; JPY repo close to BOJ policy rate plus haircut
GC repo on JGB collateral	General-collateral JGB repo rate, sensitive to BOJ JGB-supply operations
Specific-collateral repo on the corporate bond	Often more expensive; reflects bond’s borrowability and demand from short-sellers
Bank balance-sheet funding	Internal cost-of-funds (FTP); higher than repo for most banks
Prime broker financing	Hedge fund standard route via [[securities/japan-prime-brokerage-and-institutional-financing

A 50 bps negative basis can be entirely eaten by funding costs if the bond is hard to repo, the haircut is steep, or balance-sheet charges are high. The trade-economic threshold for basis trades depends on the funding stack.

Repo-availability constraints

Repo availability is the practical bottleneck. Japanese corporate-bond repo is far less liquid than JGB repo:

Bond type	Repo liquidity
JGBs (on-the-run and benchmark)	Deep GC and specific repo; sensitive to BOJ operations
JGBs (off-the-run)	Less deep; episodic specials
Japan IG corporate bonds (large issuers)	Limited bilateral repo; mainly dealer-bilateral with prime brokers
Japan IG corporate bonds (small / mid issuers)	Very limited repo; positions held outright
Japan high-yield corporate bonds	Effectively no repo; held outright on balance sheet

For most Japan corporate basis trades below Tier 1 issuer size, the bond leg is funded via prime broker financing rather than direct repo. This adds cost and reduces scalability versus US dollar or euro IG basis trades.

Basel III leverage ratio

Long-cash-short-CDS positions inflate the leverage exposure measure on bank balance sheets even when net credit risk is hedged. Post-2010 leverage ratio enforcement reduced dealer appetite to warehouse large basis positions, structurally widening basis when stress hits.

Risk-weighted assets (RWA)

Component	Effect
Bond leg	Generates banking-book or trading-book RWA based on issuer rating
CDS leg	Generates counterparty credit risk RWA on CDS counterparty
Net hedging recognition	Limited under standard approach; better recognition under internal-model approach (subject to FRTB)

Even with credit risk hedged, RWA consumption discourages dealer warehousing.

Single-counterparty credit limits

US and EU rules limit single-counterparty exposure. CDS protection-bought against the same counterparty as a cash-bond issuer creates double exposure that must be netted carefully.

Liquidity Coverage Ratio (LCR)

The bond leg generally counts toward HQLA if it qualifies as Level 2 HQLA; the CDS leg does not. Some basis trades therefore are LCR-neutral or positive for the bank, depending on bond characterization.

Japan-specific market structure post-2010

Period	Market structure shift
2010-2012	Migration to central clearing for CDS (JSCC and ICE Clear Credit); LIBOR-OIS basis spillover effects
2013-2015	BOJ QQE launch; massive JGB buying program reshaped repo specials and the JGB curve
2016	BOJ NIRP and yield-curve-control (YCC) introduction; 10Y JGB pinned near zero
2017-2019	Stable YCC era; tight Japan IG corporate spreads and narrow basis
2020 Q1	COVID credit shock; widespread negative-basis blowout across IG corporate names globally including Japan
2020 Q2-Q4	BOJ corporate-bond purchase facility and dealer balance-sheet recovery; basis partially normalized
2022-2023	BOJ YCC adjustments (widening tolerance band, then ending YCC); episodic JGB curve volatility and corporate-bond repricing
2024-2025	Continued narrow basis on most Japan IG names; episodic widening on financial names during global bank-stress events

The structural picture: dealer warehousing capacity is smaller than pre-2008, while episodic credit-risk-off events still occur. Basis trades therefore exhibit lower-frequency but higher-amplitude dislocations.

COVID Q1 2020 basis blowout

The Q1 2020 COVID episode triggered the largest negative-basis dislocation in Japan IG corporate credit since 2008-2009:

Stage	Pattern
Late February 2020	Risk-off begins; CDS spreads widen modestly faster than cash bonds
Early March 2020	Forced cash-bond selling by ETFs, dealers, and ALM portfolios; cash spreads blow out faster than CDS
Mid March 2020	Cash-bond bid-ask widened sharply; some bonds nearly untradeable; CDS remained quoted
18 March 2020	Negative basis on benchmark Japan IG corporates reached multi-year wides
Late March 2020	BOJ announced corporate-bond purchase facility and ETF buying; dealer balance-sheets stabilized
April-June 2020	Basis normalized; long-cash-short-CDS trades opened in March returned strong P&L
Q3-Q4 2020	Basis settled back to narrow normal range

The episode mirrored the 2008-2009 pattern but lasted weeks rather than months. The lesson for basis-trade practitioners: extreme dislocations are short windows requiring committed capital and tolerance for short-term MTM pain.

2022-2023 BOJ YCC adjustments

Event	Effect on basis
December 2022 BOJ widens YCC tolerance band (0.25 → 0.50 percent)	10Y JGB yield jump; Japan IG corporate cash bonds repriced; CDS lagged; basis temporarily widened
2023 incremental YCC adjustments	Episodic repricing; smaller dislocations
July 2023 BOJ formally relaxes YCC	10Y JGB upper limit effectively raised to ~1.00 percent; cash spreads widened; CDS responded with lag
March 2024 BOJ formally ends YCC and NIRP	JGB curve normalized; cash corporate bonds repriced; basis normalized over weeks

The pattern: cash-bond repricing leads CDS repricing because cash bonds are more directly held by yield-curve-sensitive portfolios (life insurers, banks). CDS spreads adjust when traders push synthetic positions to match cash, with a lag determined by dealer balance-sheet capacity.

Investor profile

Investor	Basis-trade profile
Convertible-arb / credit funds	Active basis-trade users; tolerance for MTM volatility
Multi-strategy hedge funds	Selective use, often paired with capital-structure-arb trades
Bank dealer desks	Limited prop basis trading post-Volcker / equivalent; mainly market-making warehousing
Insurance / pension	Generally not active basis traders; focus on outright cash bond holdings — see [[insurance/japan-life-insurance-alm-overview
Sovereign wealth funds	Episodic basis-trade allocations during major dislocations

Trade sizing

Basis trades are scaled by:

Constraint	Practical effect
Repo / financing capacity for the bond leg	Caps long-cash size
CDS market liquidity for the short-protection leg	Caps short-CDS size at matched name and tenor
Counterparty CSA terms (margining, eligible collateral)	Affects funding cost on the CDS leg
Single-name credit-event tail risk	Even with hedge, gap-risk during credit event
Regulatory capital cost	Bank-owned positions consume RWA / leverage

A 100 million USD-equivalent negative-basis position in a Japan IG name is large; 500 million is very large; multi-billion is rare and concentrated in the most liquid names.

Roll dynamics

If holding to bond maturity is not feasible, CDS protection must be rolled (5Y CDS → next 5Y series, etc.) at each roll. Roll-down P&L plus index-roll basis adds noise to the headline trade economics.

Sources

BOJ: monetary policy framework materials, JGB market operations data, corporate-bond purchase facility documentation.
MOF: JGB issuance and benchmark curve materials.
FSA: derivatives clearing regulation, post-crisis CDS reforms, follow-up council materials.
ISDA: CDS Definitions and Determinations Committee materials.
JSCC: CDS clearing service materials.
BIS: semi-annual OTC derivatives statistics including Japan reference-entity CDS.
JSDA: member-firm and market structure materials.