The NTV Model for Total Factor Productivity

Andrew Smithers

Published: June 2019

The consensus model for total factor productivity is unsatisfactory; the alternative, non-technology variables (‘NTV’) model resolves the objections to it and should therefore be preferred by economists. The key objections to the consensus model are that it is untestable, its assumption about corporate behaviour is falsified when tested and, for the accounting framework to function, the labour/capital ratio has to be as flexible on old capital as it is on new: an assumption which seems most unlikely. The differences in the results are non-trivial and the NTV model has positive implications for economic policy by showing how they could be changed to boost growth. NTV comprises all the variables, other than changes in labour and technology, that determine the level of investment and the capital stock. Changes in NTV are the net impact on the incentive to invest resulting from changes in the individual constituents, which are profit margins, the cost of equity, the cost of debt, leverage, corporation tax and the hurdle rate, which is the minimum expected return on equity needed to make new investment worthwhile in the opinion of management. The consensus model assumes that investment is partly determined by changes in the cost of capital while ignoring the impact of changes in the cost of equity and debt and in leverage. I show that this assumption is unjustified and why it is preferable to use NTV.

Download Paper in PDF format