To What Extent Has the Use of Monetary Policy Manipulated the Political Business Cycle?

1.0 Introduction

Get your paper done on time by an expert in your field.

Order Now

Understanding the causes of business cycles with political intervention and the factors that sustain these fluctuations has always remained an attractive subject matter for macroeconomists. The political business cycle is a theory that political influences cause fluctuations in macroeconomic cycles. It was studied empirically in Kramer’s (1971) study of economic determinants of U.S. congressional voting, followed by Fair (1978) who looked at presidential elections from 1916 to 1976. Around this early time and arguably a pioneer of the political business cycle, was Nordhaus (1975), who arrived with his innovative model from his paper The Political Business Cycle. The Nordhaus model has come to be an extremely successful model, still being used today as analysis. Shortly after was Hibbs (1977) with his model of partisan policymakers. He believed in policymakers having different macroeconomic goals with a large emphasis on partisan differences. A large influence for all of these papers was the 1972 presidential election in the United States, where Richard Nixon was accused of pre-election manipulation. Following this early work came more theoretical research on forming expectations of future models with both opportunistic and partisan fiscal models by Rogoff (1990) and Rogoff and Silbert (1988). Rogoff (1988) called Nixon “the all-time hero of political business cycles”. However, on the empirical side came Alesina (1987, 1988) with extensive work testing the original and subsequent models, looking for empirical evidence of political determinants of business cycle activity and partisan post electoral cycles.

Table 1: Writers Interpretation of Voters Assumptions and Policy Maker Motives

2.0 Literature Review

This review aims to critically assess all existing literature on the political business cycle and focus on the different interpretations economists have on the topic, with the purpose of emphasising any areas of the that require progress through further studies.

2.1 Backward-Looking Models

Opening with Nordhaus’s (1975) model, this original model was based around monetary policy being the driving force and assuming the voters were backwards looking, so they only focus on what has happened in the past instead of what will happen in the future. It’s driving theory was “expansionary monetary policy led to a temporary increase in economic activity, followed with a lag, by an increase in inflation” (Drazen, 2000). The model was to show that if voting was based on economic performance and inflation was backward looking, it would be optimal for someone who controlled monetary policy to induce an inflation-unemployment cycle with a boom just before an election and recession after. This gives the deception that the economy is stable and hitting targets, benefitting the re-elector, when they know it will most likely change following the election. This economic manipulation is thought to occur for every party trying to stay elected for another term and can be useful for them to satisfy voters that they are still able to do a good job. Nordhaus’ model came with a number of assumption to help create and understand the cycle. The key assumptions were: the political system only contain two parties which have to have compete policy divergence, both parties are out to maximise political profit instead of the ideological economy, the politicians are opportunistic non-partisan, the timing of elections are exogenous, individual voters are identical in wanting low inflation and unemployment and the most important assumptions are, policy makers control the level of unemployment by manipulating aggregate demand with monetary and fiscal policies, and expectation of unemployment and inflation are formed adaptively by agents that are backward-looking. Without the assumption that the model is backward looking the model would not work and the voters outcome would be very different. It also separates this model from other models that will be discussed that have forward-looking assumptions.

Nordhaus’s (1975) model has been tested many times empirically with econometric tests of the monetary opportunistic political business cycle, for both economic outcomes and policy instruments. It has been mostly tested with an autoregression of economic performance with a small set of economic variables and political dummies. For example, Alesina and Roubini (1992) tested the Nordhaus model on quarterly data and in 1997 Alesina, Roubini and Cohen (1997) used a dummy variable in an election quarter and the election quarters beforehand, to measure economic activity with growth rate and unemployment. However, Stigler (1973) “concluded that congressional election results are not affected by economic fluctuations” (Drazen, 2000).

The model in figure 1 is the writer’s representation of the Political Business Cycle as an adaption of the original Nordhaus (1975) model and how an opportunistic non-partisan politician, who understands the economy and voters’ preferences, can manipulate and control fluctuation in the economy to suit their re-election preferences.

Figure 1: Writers representation of the Political Business Cycle as an adaptation of Nordhaus’s (1975) model.

Period 0:

Preferred position for the economy.

$u_{0} = u^{t} π_{0} = 0$

$v_{0} = 40 % of vote$

Period 1: Pre-Election Boom

Policy to inflate economy by using monetary policy to inflate economy. This causes investment and consumption to increase, with addition of the multiplier effect, output increases and unemployment falls.

Position of economy moves to point 1.

$v_{1} = 60 % of votes$

and wins election.

Period 2: After Election

$SRPC π^{e}$

shifts up to
$SRPC π^{e} = π_{1}$

$π_{2}^{e} = π_{1}$

Economy moves from point 1 to 2, leaving inflation at the same point of
$π_{1}$

and unemployment returns to u.

Period 3:

Economy is not able to sustain high levels of high inflation output with the increased unemployment. The economy moves from point 2 to 3. The economy cannot sustain the lower level of inflation at
$π_{0} = 0$

and
$u_{0} = u^{t} - 1$

and combined with the government increasing its interest rates back to where they were before the election, the
$SRPC π^{e} = π_{1}$

moves back to point
$SRPC π^{e}$

. This leaves the economy in a worse position than it was before the election. However, it will slowly move back to point 0 and the re-elected government will stay in power.

This cycle has been widely thought to be a common pattern, however there are limitations. There have been many econometric tests of the political cycle using Nordhaus’s (1975) model, providing very little support. McCallum’s (1978) study of unemployment fluctuations before elections held no serious support and Faust and Iron’s (1999) paper with more advanced techniques came to the same conclusion. They recorded the GNP growth of the President’s term in the U.S. by quarter from 1948 to 1998 and found no real pattern or unusual fluctuations that would suggest manipulation. It should also be noted that the model assumes the president controls monetary policy, which should really be controlled by the independent federal reserve. However it is argued that monetary policy is largely controlled by presidential influence. Another more critical problem is the condescension of believing voters behaviour is irrational. Voters do not simply vote for the way they expect inflation to be and Nordhaus’ model simply has to ignore manipulative opportunistic policy makers. Finally a more ambiguous but important point is the opportunistic model ignores all fiscal pressures in the political business cycle and arguably pre-election fiscal policy can play a very large role in policy manipulation.

Nevertheless, Nordhaus’s (1975) model is a great effort of determining the decision making of non-partisan politicians. However, this model comes with many assumptions and breaking the economies fluctuations down to solely political manipulation is very difficult to prove, with there being an indefinite amount reasons that an economy’s inflation and unemployment may fluctuate. Nonetheless, you cannot deny that the motivation is there for politicians to try and help their political campaigns and there is proof from Kramer’s (1971) original research, all the way to Rogoff and Silbert (1988), that the political business cycle and the Nordhaus model help economists understand more and more about business cycles and why they behave the way they do.

A few years after Nordhaus’ (1975) model came the monetary-partisan model by Hibbs (1977). It was first introduced to explain the difference in interests of different parties.

The basic partisan model leads with the observation that left-wing and right-wing parties have different preferences on economic issues and consequently will have different macroeconomic objectives they want to achieve. In terms of their macroeconomic objectives they both have different opinions over the levels of unemployment and inflation and the targets of unemployment and inflation. The partisan-models main assumption consisted of: the political system only contained two parties, each political party had a different ideological agender, the politicians are partisan, individual voters are different in the outcome they want in terms of inflation and unemployment, expectation of unemployment and inflation are formed adaptively by agents that are backward-looking. You can see how Hibbs has taken a lot of inspiration for his model from Nordhaus, using similar framework to build upon but just expanding it to parties being partisan and wanting different outcomes.

Figure 2: Writers representation of the Operations Traditional Partisan Model as an adaptation of Alesina, Roubini, and Cohen, (1997).

Figure 2 shows the intended targets of where lef

Hibbs studied post-war patterns of macroeconomic policies and the results accompanying with left and right-wing governments in capitalist democracies. Hibbs’ paper argued that the preferences of lower income groups are “best served by a relatively low unemployment, high inflation macroeconomic configuration” whereas comparatively the higher income groups preferred a “high unemployment, low inflation configuration” Hibbs (1977). This theory was backed up by data supporting the claim with aggregate data on 12 west European and north American nations, revealing that a high unemployment, low inflation political cycle was adopted by rightist parties and a low unemployment, high inflation political cycle was adopted by nations with regularly leftist parties. Hibbs’ time series analysis of post-war unemployment data from great Britain and the united states shows labour and democratic parties drive down unemployment rates and upwards by conservative and republican governments. This conclusion aligns with Hibbs’ theory of the partisan political cycle and that governments pursue policies that are of their followers preferences. Hibbs again followed up more theory with his (1994) paper Partisan Theory After Fifteen Years.

Table 2: Primary Political Goals of Political Parties in Developed Countries

Source: Hibbs (1977)

Table 2 represents each parties political preferences and according to Hibbs, the tables was obtained and surveyed by experts in eight industrial nations confirming the choices of the different parties in his model.

A main criticism of Hibbs’ original model was that electoral uncertainty drives the model and the more uncertainty leading up to an election, the greater the magnitude of the cycle. The model also relies on mistaken expectations of what the voters desire in terms of policies. Additionally, although the partisan theory relies on the different partisan politicians having different opinions of what inflation should be, on average Democratic parties tend to have lower average inflation than Republican parties in the first half of their terms. But, in the second half, inflation is rising for Democratic parties and falling for Republican parties. Finding reported by Faust and Irons (1999). Consequently, inflation data does not support a classic partisan model. These criticisms is why further research was conducted on forward-looking partisan models by Alesina (1987) and Alesina, Roubini, and Cohen (1997).

2.2 Forward-Looking models

A decade later Alesina (1987,1988) introduced the rational-partisan model which introduces rational expectations into the PBC and suggests election outcomes may cause temporary changes in real macroeconomic variables. He introduced this theory due to the large criticism that the original model was based on an “exploitable Phillips curve” (alasina 1987). In this model only inflation shock can effect output, which is why it is called a rational-partisan model. The rational-partisan model is similar to the three equation model by Nordhaus, as it retains the expectation-augmented Phillips curve and changes two other components. However, the motive for the policy maker is different to the Nordhaus model as they are partisan, with no opportunistic motives and hence no aspiration to influence outcomes. Alesina also divides a governments time in office into two periods and makes the assumption that there is an election every other period. The other key difference Alesina makes compared to both Hibbs and Nordhaus is that he replaces the assumption of adaptive expectations with rational expectations.

One assessment of this model that (Drazen 2000) brought up was that the rational partisan model raised many questions regarding the fundamental driving forces in the initial theoretical structure. Alesina’s argument is that nominal wage contracts are signed at discrete intervals, “where nominal wage increases reflect rationally anticipated inflation at the time the contracts are signed, so that surprise inflation between contract dates can have real affects when agents are rational” (Drazen 2000). However, Rogoff (1988) explains that elections are a big factor in fluctuations due to the outcome of an election not being known. But, on the other hand everyone knows the exact election date. So the size of the changes in inflation and unemployment the model is meant to explain are arguably too large and to combat it there ought to be a sizable utility payoff to exclude the doubt that causes the fluctuations. To stop this the old contracts should expire and the signing of new contracts should wait until after the election in finished. Garfinkel and Glazer (1994) presented empirical evidence showing that labour contracts should wait until after the election if they are to be signed within the two years leading up to it. This could mean the key driving force is exogenous. Also there is a positive correlation between “electoral surprise and the size of post electoral movements in real economic activity”. However, Hibbs (1992) argued this point is invalid with as it is inconsistent with data from the United states.

So far the review of monetary models in the PBC have a theoretical and empirical issues. With each of the three models falling short with underwhelming support with data. The partisan, opportunistic and rational-partisan model all rely similarly on Phillips curve which is manipulated in the economy.

A key criticism of all the models assessed so far is they all have narrow assumptions of an economy that is not affected by fiscal policies. But the problems with the conventional political business cycle models go beyond the classic Phillips curve exploitation, when you have to consider simple government pre-election mind game manipulation. For example in most cases governments try and make themselves look better, just by lowering taxes or increasing spending, just to make them look good; according to Morse, E. and Tufte, E. (1978). Also, Keech, W. and Pak, K. (1989) found benefits of an electoral cycle in the united states between 1961 and 1978, but mentioned since then it has arguably disappeared. This is the most common empirical characteristic of the political cycle. Why should voters prefer a politician who is just disrupting fiscal flows for political gain? Furthermore, why should these actions prove the politician will do a good job over time?

This is where Rogoff’s (1990) and Rogoff and Silberts’s (1988) rational opportunistic model on elections and macroeconomic policy cycles become valuable. They argue that electoral cycles with macroeconomic policies derive from temporary information asymmetry, with variable such as; taxes, government spending, deficits and money growth. They believe the government has indicators of its performance before the voter does. Therefore, prior to the election periods the parties have incentive to signal they are performing well to voters before they know. This gives a sense of confidence and can change the electoral cycle in macroeconomic policy. However, there model does not strictly provide a rationale for an electoral cycle in unemployment. Possibly the most important reason for developing a model of political budget cycles is to enable a view of whether there are welfare implications during the election years and to see if political budget cycles cause negative consequences.

Following Rogoff’s work, Drazen (2000) introduced the Active-Fiscal, Passive-Monetary (AFPM) model. His model was based on Rogoff’s (1990) model on political budget cycles, however including monetary policy that is controlled by a separate monetary authority. Drazen’s model differs from other models covered so far due to the political cycle not only controlled by one single authority. The incorporation of an independent monetary authority with elected officials who influence fiscal policy create a different approach to the political cycle, which none of the previous models have incorporated. This more advanced model including more than one single authority controlling macroeconomic policy relates to more recent years in the UK economy, as they introduced the central bank in 1997 to independently control interest rates.

The AFPM model shows how different politicians differ in their capability to provide public goods. Their ability, however, is not directly observed by the voters. During the lead up to an election, the politician may distort the provision of public goods to make themselves look superior. The independent monetary authority on the other hand wants stable output and inflation, but the target levels of the independent monetary authority differ to the politician and the public. Therefore, the politician influences the fiscal authority and tries impress voters by increasing government consumption, forcing the independent monetary authority to increase interest rates, to level the fiscal effects on output. This fluctuation implies there should be a political cycle in both monetary and fiscal instruments but no cycle in output.

2.3 Empirical data analysis

The tests reported by Nordhaus (1975) showed his hypothesis was correct for three of the nine countries in the study, with a very strong case in the United States. The remaining six countries were more or less neutral, suggesting the evidence should be regarded as fairly suggestive. In contrast a paper published in the same month by Sargent and Wallace (1975) suggests otherwise. This paper includes detailed development of a hypothesis that suggests governments cannot, even if they try, repeatedly build booms during an electoral period. Prices may be high, but output and employment will be unchanged. An empirical test was carried out by McCallum (1978), in which he tested to see who’s proposed hypothesis was right. McCallum concluded that the political business cycle theory put forward by Nordhaus was unfavourable. However, Sargent and Wallace’s theory, also did not show particularly strong evidence either. It is possible that the electoral variables provide no explanatory power in the various tests for the simple reason that governments cannot manipulate the Phillips cure.

2.4 Concluding Remarks

To conclude this literature review, the debate of the political business cycle is clearly unfinished; with theories, models and empirical evidence coming from several different perspectives. It is evident that there is more research on the opportunistic approach, however this is most likely due to Nordhaus’ (1975) model being the first and most recognised model in the theories of the political business cycle. Furthermore, with there being lack of research in a time where there is an independent monetary authority controlling the level of interest, shows the debate of the political business cycle needs to be re-evaluated.

3.0 Project Proposal

Following the lacking literature and data collected in a post independent monetary authority period, this project will assess and evaluate Monetary policy in the political business cycle. More specifically with an analysis on unemployment, inflation and interest rates controlled by an independent authority. The academic motivation for this project is to understand whether the introduction of independent monetary authority has stopped the exploitation of a political business cycle, or whether it is still present. The main objective is to analyse these variables with UK data from 1979 up until the present day, to look at a more recent pattern and to asses if there is a difference in data following the introduction of the central back in 1997.

A Key motivation for choosing this particular area is that the debate seems unproven. Apart from the more well-known economists and their models the genuine empirical evidence does not show much promise, with each conclusion seeming to be disproven by every next theory being introduced. Each key answer comes with a list of unrealistic assumptions, which is enough to the next researcher to be deemed not enough evidence. It is also worth noting how there is a severe lack of new research being developed for this important area and since the year 2000, all research in this area has stopped completely. This gives this project a chance to see whether the political business cycle has changed or whether same conclusions are being met.

4.0 Methods and Methodology

The method of research will be a time series analysis of the variables unemployment, inflation and interest rates. The research will involve a quarterly breakdown of quantitative data to pin-point the changes in each period and to see whether in electoral periods there is a pattern to be observed. The motivation for a time series analysis comes from the lack of past research with a timeline. Also, the use a time series will show long and short-term trends which may occur within the series.

By completing a quantitative analysis, economists and researchers can improve their understand of the political business cycle and the relationships between each variable. The current previous literature needs a more up to date assessment of the political cycle with improved data and means of empirical analysis.

The reason for this methodology is to compare the different periods within the time series and to find the whether or not a political business cycle is present.

Unfinished

5.0 Skills

As this study is using a time series analysis, it will require the researcher to possess a proficient knowledge of analytical skills. During my second year of university I have studied modules involving the use of Econometrics, of which I have learnt to use programs that involve data analysis, such as Gretyl and SPSS. Although it has been a few terms since I have used these programs I feel confident after a few refreshers that they will be the best assistant to help answer this projects questions.

The proposed methods are all standard with within Econometrics, so if I feel I need to develop my skills further I can take out books from the library such as Dougherty, C. (2016) textbook, Introduction to Econometrics, as well as using in program tutorials on Gretyl and SPSS to develop the skills for my data collection process. This initial development of skills will take place alongside the data collection process, so the skills required are learned early.

7.0 Bibliography

Kramer, G. (1971). Short-Term Fluctuations in U.S. Voting Behavior, 1896–1964. American Political Science Review, 65(01), pp.131-143.
Fair, R. (1978). The Effect of Economic Events on Votes for President. The Review of Economics and Statistics, 60(2), p.159.
Nordhaus, W. (1975). The Political Business Cycle. The Review of Economic Studies, 42(2), p.169.
Hibbs, D. (1977). Political Parties and Macroeconomic Policy. American Political Science Review, 71(04), pp.1467-1487.
Rogoff, K. (1988). [Macroeconomics and Politics]: Comment. NBER Macroeconomics Annual, 3, p.52.
Rogoff, K. and Sibert, A. (1988). Elections and Macroeconomic Policy Cycles. The Review of Economic Studies, 55(1), p.1.
CUKIERMAN, A. and MELTZER, A. (1986). A POSITIVE THEORY OF DISCRETIONARY POLICY, THE COST OF DEMOCRATIC GOVERNMENT AND THE BENEFITS OF A CONSTITUTION. Economic Inquiry, 24(3), pp.367-388.
Drazen, A. (2000). The Political Business Cycle after 25 Years. NBER Macroeconomics Annual, 15, p.75.
Alesina, A. and Roubini, N. (1992). Political Cycles in OECD Economies. The Review of Economic Studies, 59(4), p.663.
Stigler, G. (1973) General economic conditions and national elections. American Economic Review 63:160-167.
Alesina, A. and N. Roubini, with G.D. Cohen. (1997). The Political Cycles of the macroeconomy, Cambridge Mass: The MIT Press.
McCallum, B. (1978). The Political Business Cycle: An Empirical Test. Southern Economic Journal, 44(3), p.504.
Faust, J. and Irons, J. (1999). Money, politics and the post-war business cycle. Journal of Monetary Economics, 43(1), pp.61-89.
HIBBS, D. (1994). THE PARTISAN MODEL OF MACROECONOMIC CYCLES: MORE THEORY AND EVIDENCE FOR THE UNITED STATES. Economics and Politics, 6(1), pp.1-23.
Hibbs, D. (1979). The Mass Public and Macroeconomic Performance: The Dynamics of Public Opinion Toward Unemployment and Inflation. American Journal of Political Science, 23(4), p.705.
Hibbs, D. (1992). Partisan theory after fifteen years. European Journal of Political Economy, 8(3), pp.361-373.
Morse, E. and Tufte, E. (1978). Political Control of the Economy. Foreign Affairs, 56(4), p.879.