Imagine you're tracking your net worth on a spreadsheet. Every month you add your savings — income minus spending — and expect the total to match your brokerage statement. But it never does. The statement keeps climbing faster than your deposits can explain. The gap isn't an error; it's capital gains, asset revaluation, and market repricing doing the heavy lifting. That's the core mechanism this paper is about: the disconnect between what gets saved and what capital actually does. Getty and Tkachenko's committed claim is blunt: net saving — the standard macroeconomic channel through which economists say capital accumulates — has contributed little or nothing to aggregate capital growth at the global scale since at least the 19th century. Their earlier 2025 paper made this case with 94-nation data from 1980 to 2021, finding a ceiling of 6.4% possible contribution from net saving. Extending the dataset backward to the earliest available year for each country tightens the noose: the margin drops to 1.6%. The method is straightforward accounting-identity analysis. They take reported or reconstructed series for market-value capital stocks and net saving flows, then ask: how much of the observed change in capital can be attributed to net saving versus other sources — primarily revaluation (unrealized capital gains, price-level shifts, and market repricing of existing assets)? The paper leans on the Penn World Table and related historical reconstruction datasets, which are the standard source for long-run national accounts but carry well-known limitations in pre-1950 data quality. The integrity picture is mixed. This is not a causal identification paper — there's no instrument, no natural experiment, no randomization. It's a descriptive accounting exercise at the macro level, which means the result is only as strong as the national accounts data underlying it. Pre-20th-century capital stock reconstructions are rough estimates, and the authors don't stress-test sensitivity to data quality regimes. The comparison is internal (their own earlier paper's 6.4% vs. the new 1.6%), not against an independent team's replication. On the other hand, the Penn World Table is a community benchmark, and the finding is robust across 94 countries over 40+ years in the modern portion. The deeper argument — that capital growth is driven by revaluation and that cash flow concepts deserve a larger role in economic analysis — places this paper squarely in a long-running heterodox challenge to the neoclassical growth model. Standard Solow-Swan and Ramsey models treat saving as the engine of capital accumulation. If saving contributes almost nothing to capital growth at the aggregate level, the transmission mechanism in those models is broken or at least misspecified. This is a big claim, but the paper is more manifesto than proof: it proposes an alternative explanation and argues for it, without building a formal model that could be tested against alternatives. The milestone question is revealing. The authors don't specify what number or test would settle the argument — there's no threshold like 'if saving explains less than X% at Y confidence with Z controls, the Solow channel is falsified.' The paper narrows a margin (6.4% → 1.6%) but doesn't define what would constitute a kill shot. For the field to move, someone needs to build a structural model where revaluation is endogenous and saving is a residual, then test it against standard growth accounting in a pre-registered framework. The obvious next experiment is a causal identification strategy — an instrumental variable or natural experiment that exogenously shocks saving rates without affecting revaluation, or vice versa. The authors don't attempt this, and the honest read is that clean instruments for aggregate saving at the national level are extremely hard to find. This isn't computational budget — it's a conceptual bottleneck. The paper is doing the descriptive groundwork that would motivate such a study, but the heavy lifting remains undone.
