For productivity analysis, the appropriate measure of capital input is the flow of capital services, i.e. the flow of productive services that can be drawn from the productive capital stock. This productive capital stock is the cumulative stock of past investments in capital assets adjusted for losses in their productive capacity (or efficiency) and retirement (Schreyer, Bignon and Dupont, 2003[2]). Conceptually, capital services should not be confused with the value of capital that is measured by the net wealth capital stock. For example, the capital services provided by a taxi relate to the number of trips, distance driven, and comfort of the taxi, rather than the market value of the vehicle, which would instead relate to the net wealth capital stock concept. These services are estimated using the rate of change of the productive capital stock of different capital goods and aggregated using rental prices or user costs shares as weights (as opposed to market price shares used to aggregate net wealth capital stocks).
As part of the revamp of the OECD Productivity database, the method for measuring capital services has been refined. The new estimates of capital services offer a more detailed breakdown by asset type and industry and are fully aligned with national accounts. This consistency stems from the depreciation rates used in calculating asset user costs and return rates are fully in line with those applied by national statistics offices in deriving net capital stock. In addition, this way of calculating capital services does not require the extension of gross fixed capital formation (GFCF) series or strong assumptions to estimate initial capital stocks. This differs from the previous version of the database, which assumed an identical depreciation rate for a given asset in the same industry across countries (OECD, 2025[3]).
MFP growth is measured as a residual, i.e. by the part of GDP growth that cannot be explained by the contributions of labour and capital inputs to GDP growth. Traditionally, MFP growth is seen as a measure of technical change but, in fact, technical change can also be embodied in factor inputs, e.g. improvements in design and quality between two vintages of the same capital asset. In practice, MFP only captures disembodied technical change, e.g. network effects or spillovers from production factors, the effects of better management practices, organisational change and improvements in the knowledge base. Moreover, MFP picks up other factors such as adjustment costs, economies of scale, effects from imperfect competition, variations in capacity utilisation (if not captured by the capital input measures), and errors in the measurement of output, inputs and input weights. For instance, increases in educational attainment or a shift towards a more skill-intensive production process, if not captured by labour input measures (i.e. labour services) will end up in measured MFP. Moreover, tax policies can also affect the measurement of MFP. Profit shifting driven by differences in corporate tax regimes may increase reported output in low tax countries without changes in real production, resulting in inflated MFP levels (Bricongne, Delpeuch and Lopez-Forero, 2023[4]). Therefore, a reliable measure of MFP requires accurate data on both outputs and inputs .
In Figure 2.1, national accounts figure on hours worked for Austria, Estonia, Finland, Greece, Latvia, Lithuania, Poland, Portugal, Sweden and the United Kingdom have been replaced with estimates obtained with the OECD simplified component method described in the Section on Hours Worked and Employment in Chapter 5. However, the impact of this correction on labour productivity growth rates is marginal (Ward, Zinni and Pascal, 2018[5]).
For further methodological information, consult the OECD Productivity Statistics – Methodological notes at www.oecd.org/sdd/productivity-stats/OECD-Productivity-Statistics-Methodological-note.pdf and the update of the sources and methodologies used to construct the new OECD Productivity Database (OECD, 2025[3]).
