An Empirical Investigation into Stock-Flow Norms

Smith, Graeme (2019). An Empirical Investigation into Stock-Flow Norms. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000f1e2

Abstract

The effect of a stock in a flow system is to absorb shocks to the flow as can be seen in many applications, from capacitors to flood plains. The ratio of the stock to its related flow may form a stationary time series, in which case its stable value is considered to be a norm. Such norms provide stable reference points in economic dynamics and have been associated with the New Cambridge school and the stock-flow consistent (SFC) modelling approach.

Using national accounts data for the US from 1960 to 2016, some common stock-flow ratios are investigated for stationarity, focussing in particular on those relating net financial assets to income for the sectors of an open economy. If sectoral balances are roughly stable, the wealth-income ratio will be stationary. It is found that, due to increased volatility over recent decades, stationarity is quite rare.

The norms figure in a dynamic process in which the fiscal stance and trade performance ratio, here termed the flow ratios, act through a multiplier to drive national income which is stabilised by the stock-flow norms. The process leads to convergence to the dominant flow ratio under the action of partial adjustment processes; which flow ratio is dominant depends on characteristics of the particular economy. Simulation is used to study the path of convergence and flow ratio dominance. The simulations are found to exhibit complex systems behaviour under certain circumstances.

A three sector SFC model is populated with the US data and estimated using a Johansen cointegrating VECM; interpreted as partial adjustment processes, the cointegrating relations for each sector yield estimates of the partial adjustment parameters and the stock-flow norms. This econometric approach emulates the Hendry data-first methodology whereby a pure data model captures the DGP in which postulated theory models are capable of interpretation.

Viewing alternatives

Download history

Metrics

Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions

Export

About