Total factor productivity measures output growth not accounted for by growth in measured labor, capital, and other production inputs.
Total factor productivity (TFP) measures the portion of output growth not accounted for by growth in the measured production inputs. It compares real output with a combined input measure that may include labor and capital or, in more detailed industry models, capital, labor, energy, materials, and purchased services.
TFP is also called multifactor productivity (MFP). The terms are often interchangeable, but the exact output boundary, input set, and methodology must be checked before comparing estimates.
For a value-added model with capital and labor, a common growth-accounting expression is:
where:
A is total factor productivity;Y is real output;K is capital input, preferably capital services;L is labor input; ands_K and s_L are the income or cost shares attributed to capital and labor.Under the simplified assumptions commonly used in growth accounting, the shares sum to one. Official estimates use more detailed index methods, time-varying weights, multiple asset and worker groups, and carefully matched output and input boundaries.
An industry-level gross-output model may use a KLEMS input set: capital, labor, energy, materials, and purchased services. A TFP estimate from that model is not directly interchangeable with a value-added TFP estimate based only on capital and labor.
Assume real output grows by 4.5%. Growth in measured capital input contributes 1.0 percentage point and growth in labor input contributes 1.5 percentage points. The residual is approximately:
The model therefore reports about 2.0% TFP growth. That result does not prove technology alone raised output by 2.0%. Better organization, changing utilization, economies of scale, shifts toward more productive establishments, unmeasured intangible capital, and data error can also enter the residual.
| Feature | Labor productivity | Total factor productivity |
|---|---|---|
| Basic comparison | Output per labor hour | Output relative to combined measured inputs |
| Capital contribution | Included in the outcome but not separated | Capital input is explicitly accounted for |
| Typical frequency | Often quarterly and annual | Commonly annual because input data are detailed |
| Main use | Timely view of output per hour | Analysis of growth beyond measured input accumulation |
| Main interpretation risk | Mistaking it for worker effort | Mistaking the residual for pure technology |
Labor productivity can rise because workers receive more capital per hour even when TFP is unchanged. A simplified decomposition of labor-productivity growth can include TFP growth, a capital-deepening contribution, and changes in labor composition.
The measured residual can reflect a mixture of:
This is why TFP is informative but not self-explanatory. It tells an analyst what measured input growth did not account for inside a model; it does not identify a single cause without additional evidence.
Long-run growth: Input accumulation has limits. TFP analysis helps assess whether output is expanding through more inputs, more effective use of inputs, or both.
Industry comparison: TFP can reveal different efficiency paths after accounting for measured capital and labor. Comparisons still require consistent industry boundaries, price indexes, and methods.
Investment and valuation: Productivity assumptions can affect long-run margin, capacity, and growth forecasts. Company analysts rarely observe an official firm-level TFP measure, so operational proxies should not be presented as equivalent.
Policy analysis: TFP trends are used to study innovation, competition, infrastructure, market allocation, and other potential growth drivers. The residual itself does not prove which policy caused a change.
In 2021, the U.S. Bureau of Labor Statistics replaced multifactor productivity with total factor productivity in its publications. BLS describes this as a terminology change that did not alter the data or estimation methods. Other institutions and older research may continue to use MFP.
TFP is an educational statistical concept. It does not by itself provide a forecast, investment recommendation, or causal policy conclusion.