{"slug":"credit-impulse","name":"Credit impulse","family":"Macro-financial conditions","status":"active","estimate":"The quarter-on-quarter second difference of the BIS private-sector credit-to-GDP ratio for nine economies.","page":"/conditions","implementation":{"enginePaths":["engine/finweave_engine/layers/macro/credit_impulse.py"],"producerPath":"scripts/build_conditions.py","outputs":["data/parquet/conditions_series.parquet"]},"methodology":[{"path":"engine/finweave_engine/layers/macro/credit_impulse.py","lines":[11,31],"note":"compute_credit_impulse() computes the first and second differences of the credit-to-GDP ratio."},{"path":"docs/conditions_methodology.md","lines":[171,202],"note":"Defines this build's raw quarter-on-quarter second difference and distinguishes it from the original smoothed bank-credit-flow construction."}],"data":{"vintage":"2,236 quarterly observations for 9 economies from 1948-06-30 through 2025-12-31, measured 2026-08-30","tables":["data/parquet/conditions_series.parquet","data/parquet/bis_credit_gap.parquet"],"sources":[{"label":"Bank for International Settlements, total credit and credit-to-GDP gaps","url":"https://data.bis.org/topics/TOTAL_CREDIT"}]},"method":["The engine forms credit divided by GDP, differences that ratio once to obtain its quarter-on-quarter flow, and differences the flow again. The producer passes the BIS private non-financial-sector credit-to-GDP ratio as credit and a constant one as GDP, so the published value is exactly the ratio's quarter-on-quarter second difference.","This is a broad credit-to-GDP acceleration measure. It is not a smoothed bank-credit flow and it does not reproduce the original year-on-year Biggs-Mayer-Pick construction."],"equations":[{"label":"Credit ratio","expr":"ratio = credit / GDP"},{"label":"Credit flow","expr":"flow = ratio.diff()"},{"label":"Credit impulse","expr":"impulse = flow.diff()"}],"validation":["The output contains 2,236 credit-impulse observations for 9 economies from 1948-06-30 through 2025-12-31.","The producer drives the same engine function for every economy with the BIS ratio and a constant denominator, making the stored measure's second-difference interpretation explicit."],"limitations":["Raw quarter-on-quarter second differences are noisy; turning points carry more meaning than isolated quarterly extrema.","The input is broad BIS total credit to the private non-financial sector, not bank lending alone.","The reversal after the 2020 credit spike is partly mechanical and should not be read by itself as a crisis signal."],"references":[{"citation":"Biggs, Mayer, and Pick (2010), Credit and Economic Recovery: Demystifying Phoenix Miracles.","url":"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1595980"}],"endpoints":[{"path":"/api/methods/[slug]","example":"/api/methods/credit-impulse","method":"GET","returns":"Registry metadata, implementation paths, measured vintage, methodology evidence, validation, and limitations. No model observations or parquet contents."}],"lastVerified":"2026-08-30"}