![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
It is widely reported for many countries, including the UK, that income velocity has been highly variable around a declining trend in recent years. This paper advances the following hypothesis. The demand for credit and hence the broad money stock are influenced by total spending in the economy, rather than spending only on newly produced goods and services. Since total spending in the economy has generally increased relative to GDP (mainly because of asset transactions) credit and money have expanded more rapidly than GDP, with the resulting fall in income velocity. Using quarterly data from 1975 to 2008, we estimate a vector error correction with income velocity as the dependent variable and the ratio of total to GDP transactions as an explanatory variable. The results show substantial support for the hypothesis and raise (further) doubts about the information content of broad money aggregates for inflation targeting central banks.
Acknowledgements
We are grateful for the comments of two anonymous referees and the editor of this journal. Any remaining errors are our own responsibility.
Notes
1. In 2008 APACS became the UK Payments Administration. Its website is: http://www.ukpayments.org.uk
2. ‘CHAPS’ is an electronic bank-to-bank same-day payment facility available to member banks. The main feature of CHAPS is that it is fast, secure and efficient and the money is transferred the same day. Unlike other forms of payment such as cheques, CHAPS payments are irrevocable. The pricing structure is such that small payments are discouraged. Most payments involve settlements of financial transactions between banks and other financial institutions. In 2008, the average size of CHAPS payments was about £3 million.
3. Dreger and Wolters (Citation2009, 53) make the point that the sign on the wealth variable is ambiguous because higher non-money asset prices make non-money assets more attractive than money and so agents switch from money to non-money. They describe this as a (negative) substitution effect that could dominate the (positive) scale effect that follows from higher wealth. But it is not clear why higher non-money asset prices should have this effect. Higher returns on non-money assets would have a negative substitution effect, but these higher returns on non-money assets imply lower prices. Hence, when written in price terms, one would expect a positive substitution effect (and no ambiguity).
4. For more details, see Arestis and Howells (Citation1992).
5. The range of data can be easily seen by visiting the website: www.apacs.org.uk
6. For further discussion of the APACS data, including some caveats regarding its use, see Bain and Howells (Citation1991).
7. Including the CHAPS data produces a similar shape to the plot but it begins at about 20 (instead of 2) and rises more steeply to 50 (instead of 3). It then falls back to about 40.
9. The ‘range’ is some measure of volatility. We had to resort to this measure because some business cycles were too short to calculate a meaningful standard deviation. The details are available on request.
10. The order of integration for is and il is not clear cut, since the ADF (inclusive constant and trend) indicates stationarity while this is not so for the ADF (inclusive constant). The deterministic trend variable is not significant in the ADF regression. Furthermore, the correlograms for both interest rates indicate non-stationarity in levels (see the Appendix). The results of the unit root tests for inflation, and the short-term interest rate are the same as, for instance, in Hendry and Mizon (Citation1993), albeit for a different time period.
11. The trace test for the long- and short rates is 23.93 with a probability level of 0.014. The coefficient restriction (il = is ) is not rejected with ð2 = 0.522 and a probability level of 0.47.
12. Recent references are Dreger and Wolters (Citation2009) and Andrés, Lopez-Salido, and Nelson (Citation2009).
13. See also Dreger and Wolters (Citation2009), and many more.
14. We are grateful to a referee who pointed this out to us.
15. We slightly changed the start and end dates of the policy regimes. It did not make any difference to the following estimation results.
16. The dummy variable has a value of one from first quarter 1986 until fourth quarter 1995.
17. The value of the dummy variable is equal to one for 1978Q4 and zero otherwise.