<![CDATA[Bank stock valuation is often modeled as a direct function of accounting ratios, yet the literature reports unstable signs for net interest margin (NIM), return on assets (ROA), loan-to-deposit ratio (LDR), capital adequacy ratio (CAR), and non-performing loans (NPL). This review asks when these indicators become informative for stock performance and whether gross domestic product (GDP) should be treated as a conditioning state rather than a routine control. A structured integrative search of Crossref and OpenAlex identified 2,056 records published from January 2021 to July 2026. After deduplication, relevance screening, and extended metadata or abstract appraisal, 56 evidence sources were synthesized; four methodological and primary institutional sources supported reporting and construct definition. The synthesis shows that ROA provides the most stable positive signal, whereas NPL has the most consistent adverse association. NIM is positive only when asset yields reprice faster than funding costs and the margin is not produced by excessive risk. LDR and CAR exhibit trade-offs: both support intermediation and resilience, but extreme liquidity deployment or excess capital can weaken market valuation. GDP growth changes borrower quality, credit demand, margins, provisioning, and risk appetite, making the ratio–return relationship asymmetric across expansions and contractions. The review therefore proposes a state-contingent framework in which GDP moderates five bank-fundamental channels through earnings persistence, funding risk, loss absorption, and credit impairment. The framework clarifies contradictory findings and provides testable designs for bank-level research, including the Indonesian market.]]>
