[DSRP Evidence](https://dsrpevidence.org/)

# China's Local Debt Reform for Fiscal Sustainability: Cost Reduction Effect and Allocative Boundary

Yongling Wang, Lin Lu, 2026, Sustainability — Economics

Patterns: [Distinctions](https://dsrpevidence.org/pattern/distinctions), [Relationships](https://dsrpevidence.org/pattern/relationships)

## In short

The finding shows that redrawing the legal boundary of government liability for firm debt produces a relational (causal) effect on borrowing costs that appears only for firms inside the redrawn boundary and vanishes for otherwise-similar firms outside it.

## What they found (results)

Hardening the legal boundary of local governments' liability lowered local state-owned firms' cost of debt by 0.33 percentage points relative to private firms (5.5% of the mean), while central state-owned enterprises unaffected by the boundary change showed no significant shift, and investment rose without a detectable gain in investment efficiency.

## What they set out to do (purpose)

To test whether China's 2023 legal disclaimer of local governments' implicit backing for their state-owned enterprises' debt changed those firms' cost of debt and investment behavior.

## Abstract

State Council Document No. 43 and the revised Budget Law legally disclaimed Chinese local governments' responsibility for the debts of the enterprises they own. Across 22,233 firm-year observations on 1686 listed firms from 2009 to 2023, the effective cost of debt of local state-owned firms fell by 0.33 percentage points relative to private firms, 5.5 percent of its mean. The direction was not obvious in advance: withdrawing a guarantee that lenders had been pricing should have made credit dearer, while the debt swap enacted alongside the disclaimer replaced high-cost vehicle liabilities with low-cost provincial bonds and eased the balance sheets standing behind those firms. Central state-owned enterprises, whose support the reform left untouched, serve as a falsification group and show no statistically distinguishable change. The estimate passes the joint pre-trend test, survives matching, entropy balancing and twenty specification changes, and operates through the interest paid rather than the quantity of debt, which identifies a price effect. Investment rose by 0.76 percentage points of assets, but no improvement in investment efficiency was detected, and an equivalence test cannot exclude a small one. A cross-sectional pattern consistent with fiscal relief does not survive correction for multiple testing, so that channel is reported as suggestive rather than identified. Hardening the budget constraint of local governments lowered the debt servicing burden of the firms they own without directing the freed resources toward more productive use, so the sustainability gain is fiscal rather than allocative.

These researchers were not testing DSRP. The finding is theirs; the correspondence to DSRP is drawn by this site.

[Source](https://doi.org/10.3390/su18189231)