Breaking the data silos of government is necessary to build a unified national market in the new era. How data elements serve enterprise’ growth constitutes a key aspect of high-quality economic and social development. Relying on the “tax credit bank loans” platform, data elements can release financial momentum through cross-departmental sharing, thereby affecting the financing constraints of small and medium-sized enterprises (SMEs).With the help of the quasi-natural experiment of tax credit bank loans implemented one after another in several Chinese cities, and using the listed companies on the SME Board and the Second Board from 2009 to 2022 as the research sample, this paper studies the impact and mechanism of government data circulation on the financing constraints of SMEs, and conducts heterogeneous analysis based on different property rights and the level of financial resource development. The main findings are as follows. The financing constraints of SMEs in locations where tax credit bank loans have been implemented have been significantly alleviated compared with those in locations where tax credit bank loans have not been implemented, and this conclusion still holds after a series of robustness tests and endogeneity analyses. The mechanism test results indicate that, this can be achieved by reducing information asymmetry between SMEs and banks, improving principal-agent issues between SMEs and banks, and boosting the confidence of other stakeholders in the future development of SMEs. Moreover, this alleviating effect is more pronounced in non-state-owned SMEs and SMEs in areas with scarce financial resources. This paper shows that tax credit bank loans, as an example of cross-departmental sharing of data elements, contribute to the high-quality development of SMEs, which not only expands the academic research on “tax credit bank loans” and enterprise financing constraints, but also provides policy references for the Party and the State to continuously improve the institutional mechanism of market-oriented allocation of data elements.