The Shift in Bank Credit Allocation: New Data and New Findings (Bezemer, Samarina & Zhang)
A De Nederlandsche Bank study assembling a new disaggregated bank-credit dataset for 74 economies, 1990–2013/2016, documenting a 'debt shift' in bank lending away from non-financial business and toward household mortgages — the broadest-panel confirmation that modern bank credit growth is.
Overview
"The Shift in Bank Credit Allocation: New Data and New Findings" by Dirk J. Bezemer (University of Groningen), Anna Samarina (De Nederlandsche Bank), and Lu Zhang (Utrecht University) was issued as De Nederlandsche Bank Working Paper No. 559 (2017) and developed into the peer-reviewed article "Does mortgage lending impact business credit? Evidence from a new disaggregated bank credit data set," Journal of Banking & Finance vol. 113 (2020).[1][2] Its core contribution is a new disaggregated bank-credit dataset that splits domestic bank credit into four categories — "home mortgages, consumer credit, bank loans to non-bank financials, and loans to non-financial business, for 74 economies over 1990–2013" (extended to 2016 in the journal version).[1][2]
Using that data, the authors "document key trends including the shift in bank credit allocation away from traditional business lending" — the phenomenon they and the surrounding literature call the "debt shift."[1] In fixed-effects and system-GMM regressions they find "debt shift is larger in advanced economies with a stronger presence of foreign banks and higher trade," and that "financial deregulation strongly correlates with debt shift."[1]
Why It Matters for the Land-Credit Claim
This is the broadest-panel confirmation of the composition fact underpinning the wiki's finance-growth-is-land-credit claim: that the long-run growth of modern banking is, above all, the growth of lending against real estate rather than of productive business lending. Where Jordà, Schularick & Taylor's Great Mortgaging establishes the mortgage-share doubling across 17 advanced economies since 1870, and Bezemer & Hudson (2016) reproduce it on 14 OECD economies (1990–2011), this paper extends the same disaggregated-credit measurement to 74 economies, advanced and emerging, with a full public account of sources and methods — the most comprehensive dataset of its kind. It is the same author's more careful, wider-sample statement of the composition claim, and (unlike the "finance income is rent" interpretation the wiki flags as contested) the debt-shift composition finding is a documented data trend, not an interpretation.
Nuance and Limits
- The paper measures the destination of credit (mortgages vs business loans), not the land-versus-structure split of the collateral; the step that the appreciating part of that mortgaged real estate is land is supplied by the house-price literature (Knoll, Schularick & Steger), not by this dataset.
- Its regression contribution is about the drivers of debt shift (deregulation, foreign-bank presence, trade); the journal version separately examines whether mortgage-credit expansion crowds business credit, a distinct question from the composition trend cited here.
- As with the rest of the "debt shift" literature, the stronger claim that this credit income is economic rent is attributed argument, not part of the composition finding.
Bears On
- Outcome (supports): The growth of modern banking is largely mortgage credit against land — the widest cross-country dataset documenting the shift in bank credit away from business lending and toward household mortgages.
- Objection: Cycles are driven by credit, not land — the debt-shift data are common ground: the credit driving the cycle is overwhelmingly real-estate credit.
- Concept: The FIRE Sector.
See Also
- The Great Mortgaging — the long-run (1870–) mortgage-share doubling
- Bezemer & Hudson: Finance Is Not the Economy — the same author's 14-OECD composition figures and the contested rent interpretation
- Mian, Sufi & Verner: Household Debt and Business Cycles Worldwide — why the mortgage-credit expansion predicts slower growth
- The growth of modern banking is largely mortgage credit against land
Sources
- Dirk J. Bezemer, Anna Samarina & Lu Zhang (2017), "The Shift in Bank Credit Allocation: New Data and New Findings," De Nederlandsche Bank Working Paper No. 559. SSRN — used for the four-category disaggregated dataset (home mortgages, consumer credit, loans to non-bank financials, loans to non-financial business) for 74 economies over 1990–2013, the documented "shift in bank credit allocation away from traditional business lending," and the deregulation/foreign-bank/trade correlates of debt shift. Abstract fetched and quotations (under 50 words) verified verbatim 2026-07-13.
- Dirk Bezemer, Anna Samarina & Lu Zhang (2020), "Does mortgage lending impact business credit? Evidence from a new disaggregated bank credit data set," Journal of Banking & Finance, vol. 113 — the peer-reviewed journal version, extending the dataset to 1990–2016 and analysing the effect of mortgage-credit flows on business credit. RePEc record. Abstract verified 2026-07-13.