The effect of UK fiscal policy on output: a historical narrative approach

Soares Coelho Pinto de Sousa, Joao (2024) The effect of UK fiscal policy on output: a historical narrative approach. PhD thesis, Birkbeck, University of London.

Abstract

This thesis contributes to the debate on the effectiveness of fiscal policy in stimulating output, investigating how to measure that effectiveness, what it has been in the UK over the 140 years from 1879-80 to 2018-19, and under what conditions these effects might differ. This thesis makes use of original archival research to identify fiscal shocks from UK parliamentary documents, both in terms of discretionary spending and fiscal stance. Chapter 2 uses data from the Jordà-Schularick-Taylor macrohistory database to illustrate the sensitivity of multiplier estimates to modelling decisions when using the Gordon-Krenn transformation. This transformation consists of dividing output and government spending by potential output, and this choice of potential output estimation method can lead to multiplier estimates as low as -0.04 and as a high as 0.70. This huge parameter uncertainty for what is a seemingly innocuous choice and which results in very similar potential output estimates. Instead, I propose returning to estimating an output elasticity with respect to government spending and then multiplying it by a conversion ratio such as the inverse of the share of government in total output. This is not only more transparent, but it also results in narrower variation than that induced by the different methods of estimating potential output. Chapter 3 makes use of a novel dataset comprising of government spending shocks going back 140 years to estimate UK-specific multipliers. This dataset is compiled from archival research in the UK Parliament, consisting of changes between the estimate for government spending at the beginning of the financial year and the estimate in the subsequent budget of how much has actually been spent. This effectively presents a series of intra-year, discretionary spending shocks – excluding cyclical components, that is, social security and debt interest – which are unlikely to be anticipated, both given the UK’s idiosyncratic budget process and statistical testing. The results point to a cumulative multiplier of 0.44 on impact and 0.47 in the long-run, as well some evidence of larger stimulative effects of civil spending relative to military spending at short horizons. This chapter’s results also support theoretical and empirical findings of falls in household consumption in response to increases in government consumption, as well as higher multipliers in times of high slack – as measured unemployment considerably above the natural rate – but not for different regimes such as the debt-to-GDP ratio, openness to trade and exchange rate regimes. Chapter 4 augments the historical analysis of chapter 3 by adding further archival research to include tax changes since 1879-80, and combines these two strands of government policy to create a combined series of changes to the discretionary fiscal stance over the course of 140 years. This allows for the estimation of a historical impact of a 1% of GDP increase in the fiscal balance on output of -0.24% in-year and -0.38% by year 4. This chapter also provides evidence of a stronger effect of fiscal policy on output in times of high slack and of a stronger increase in household consumption as a response to a fiscal tightening in times of fiscal distress, as well as weak evidence to support asymmetric effects between expansionary and contractionary fiscal policy.

Metadata

Statistics

DownloadsShow export options

Export as XMLJSONCSV

Activity Overview

6 month trend

29Downloads

6 month trend

40Hits

Additional statistics are available via IRStats2.

Archive Staff Only (login required)

Edit/View Item