Political Parties and the Size of Government: Quantitative Approaches and Policy Analysis - Term Paper Example

POLITICAL PARTIES I INTRODUCTION Politics have always been considered as an important and integral part of the social life of human beings. This fact is believed and adhered to on the premise that the political arena is the sole venue with which human beings maybe able to develop and attain human excellence (Arendt, 1958). Being such, it is not surprising that people from different epochs and cultures have come up with means and ways with which they can perform and undertake political action in the service of the society. And the contemporary society and human beings are not different from peoples in the past in this term. However, because of current developments in the history of humanity, contemporary engagement in politics may be deemed to be a little more complicated and a lot more complex. In this regard, this paper intends to look at the dynamic interaction between political parties and size of the government in lieu of public policy making in the United States of America. This is being pursued on the supposition that political parties and the size of the government are two important factors that are basically deemed to be interconnected and interrelated as the government comes up and implements policies and laws that are intended to be beneficial to the American citizens. As such, this paper intends to address the question “ is there a correlation between political parties and government size in terms of public policies being made and undertaken by the government?” This particular question is of vital importance since it has been noted that politics and government in the United States has dramatically changed. In fact it has been claimed “not only the size of the government has changed in the past 15 years, growing markedly as a share in the GNP but also the role of government has changed as well” (Roubini & Sachs, 1989 p 109). And the changed that they are referring to is the marked attribute of the United States government – a divided government. In this light, the researcher hopes that it may be able to help in any way it can in untangling and clarifying the connection and relation between political parties and government size in public policy making in the United States of America. II UNITED STATES OF AMERICA: ITS POLTICAL PARTY SYSTEM AND GOVERNEMNT Since the Post World War II, the government of the United States have been considered as divided for the reason that “American politics has been characterized by a divided government - twenty six years of Republican Presidents and at least a Democratic House of Representatives and two years of Democratic President and a Republican Congress compared to twenty years of unified government with all but two of these years being unified Democratic government” (Brady, 1993, p 189). This is considered as something significant in American politics for the reason that political scientist are now questioning the empirical repercussions in having a divided government specially when it comes to key issues pertaining to economic policies, international relations, migration and health care. The constitution of United States has a provision, which explicitly speaks of the constitutional separation of power between the executive, legislative and judicial branch of the government. Under this particular system, each branch of the government are basically autonomous from one another but working under the ethos that there is synchronization of movements among the three branches of the government as it delivers and provides basic services to the people. Although it should be noted that each government branch is independent from one another, there is a check and balance that acts as an arbiter to the possible abuse of discretion in the use of power. For instance, there are certain economic policies, which the president cannot implement unless the Congress has enacted a law allowing him to implement it (Feldstein, 1992,p1225). As such, it has been stipulated that government services can be rendered in a more expedite manner if the President and majority of the members of the Congress belong to the same party since it is held that they share the same ideals and visions for the country. However having a unified government is not an easy task in the light of the political party system of the United States. The political parties in the United States are basically polarized between Democratic Party and the Republican Party even if the political party system of US is “multi-party system” (www.wikipedia.com.). There is a Third party, which is consisted of Libertarian Party, Constitution Party and Green Party. Also there are other smaller parties like American First, American Party, American Patriot Party and many more who do not field any presidential candidates but are holding local offices. And as stated earlier, although the political party system of America is a multi- party system, the US political battlefield is technically dominated by two political parties – Democratic and Republican –who in turn fight it out in the Congress and in the Presidential arena. What poses as a problem is the fact that for the past half century, United States government have been divided literally. If the President is a Republican as with the current case of George W. Bush, the Congress is dominated by the Democrats. And this is not something new. Since, as we have stated, this has been the situation of the American government since Post World War II. Seeing the current predicament of US government as possible cause of ‘problems some political scientists have already proffered some possible interpretations regarding the government division of US. Some have claimed that it appears that the Democrats are winning in the local front based on the perception that voters put into office people who are capable of delivering services and reducing local taxes that has no direct bearing to national economic issues. But it is noted that when it comes to the national election, it appears that voters go for Republicans on the supposition that the Republicans have a more cogent national economic policies compared with the Democrats who in supporting “locally popular programs when aggregated nationally the result is that the majority of Democratic party in the House sounds like a cacophony of special interests saleable one at a time locally but not coherently aggregated” (Brady, 1993, p190). Another interpretation being given is that it seems that the Republican can balance the left wing and the middle wing providing a more stable national political field. Although the researcher strongly holds that there is really an impending need for an explanation of this particular political phenomenon being experienced in the United States of America thus paving for empirical or theoretical explanations and hypothesis to explain the apparent government division in American. However, the researcher strongly believes that what should be noted and given much attention is the possible repercussion of the division as seen in the positions supported and lobbied by the two giant parties in the succeeding section. III. DATA AND ITS SOURCES The dataset which will be used for the analysis covers the 48 “continental” US states with in the period of 1970 to 2000. Total general spending, total spending on welfare programs, state personal income, the party affiliation of the governor and the number of republicans and democrats in the state senate and state house are the main variables contained in the data. In the political variables, the governors party affiliation takes the value of one if the the governor of a particular state in a given year is a member of the Democratic Party. While if the governor comes from the Republican party, the value that will be given will be zero implying a missing value which may likewise stipulate an independent nominee. Other political variables are employed in order to know the average percentage share of the Democrats and the Republicans in both the upper and lower state houses of Congress. Further more, the dependent variables used in the regressions are total state general expenditure as a share of state personal income (expinc); total public welfare expenditure as a share of state personal income (welfinc) and the share of total public welfare expenditure in total state general expenditure (welfexp). All the data utilised for this paper are gathered from different sources. . Data gathered from the Senate and the House of Representatives are taken from Book of the States by the Council of State Governments. Information pertinent to the party affiliation of the governor (governor dataset) are provided by the National Governors Association. The total general spending and welfare expenditure data (spending dataset) are contained in the periodical State Government Finances which is published the Bureau of the Census. While, the data from State personal income (spi dataset) for the entire period are available in the website of the US Bureau of Economic Analysis. The downloaded data are prepared in Microsoft Excel and are subsequently converted to STATA format using STAT TRANSFER 8 program. Lastly, all the data are merged into one STATA dataset (dataset_complete), which, then, can be used to estimate the regressions of interest. IV. EMPIRICAL APPROACH In order to arrive at a good estimation on the effect of political parties to government size, the following general linear regression model is applied. It should be noted that under this model, state and time fixed effects are controlled. (Government size)it = i + t + (gipit) + (housesit) + (gipit x housesit) + it The dependent variable government sizeit stands for the share of total state general expenditure in state personal income. In later regressions, it likewise represents, the share of total public welfare expenditure in state personal income. The variable gipit is a binary (dummy) variable which is given the value of one if the governor of a particular state I in a given year T is a representative of the Democratic Party. And the designated value of zero if he or she is from the Republican Party (missing values in the dataset suggest an independent nominee). In the creation and establishment of policies that are prioritised, of vital and critical importance is the majority party in the House. Being such, the variable housesit as a dummy variable corresponds to the majority party in the Upper and the Lower Houses. Thus, in the regressions I use the variables dem_housesit and rep_housesit to show who dominates the upper and the lower house, whether the majority is the Democrats or the Republicans, respectively. As it is recognised that concept of government size is not only the function of the legislature but is also the function of the state governors office, it is but essential to understand that the party who gains control over both institutions could be a decisive factor in terms of effects on government size. Thus, in order to account for such an effect, we can add to the regressions the interaction term (gipit x housesit). However it should be noted that if the interaction term is not significant, it presupposes the notion that there is no concrete evidence which may suggests that the majority partys control of both Houses has a bigger effect if the governors office is also being held by the majority party. In some of the regressions reported, I also utilised the composite index of party power as presented by Berry and Lowerly (1987). The variable party_idxit is constructed in such a way wherein the party of the governor counts for 0.5 and the dominant party in the upper and in the lower house counts for 0.25 each. Thus, the party index for the Democrats captures the degree of their control over state power. So as to reduce concerns regarding the possibility of omitted variable bias, state fixed effects and time fixed effects are employed. In the equation, i signifies state fixed effects controlling all variables that vary across states but are constant over time. State fixed effects gains more importance especially in cases where the variables are difficult or simply impossible to observe like for example different institutional or cultural background. On the other hand, time fixed effects (t) capture factors that vary over time but take on the same value for each state. Thus, controlling for instance, business cycle, which in turn can possibly affect government spending in the particular year. But as a word of caution, this assumption pertinent to time fixed effects is not to be appreciated as necessary but only as plausible. Moreover, as the panel dataset contains time dimension, there is the presence of the potential threat of autocorrelation that occurs when error terms are correlated across observations. The moment this happens, a downward bias of standard of errors is caused thereby, creating an upward bias of t-statistics. Which, then in turn, substantially affect our judgment of statistical significance of some coefficients. To prevent such an occurrence, or at least to reduce similar concerns, clustering of standard errors is used. In this particular procedure, it is assumed that the error terms within a cluster can be arbitrarily correlated, but that it should be made clear that there is no correlation of the error terms across clusters. Thus, in the estimation procedure I use clustering on states. This means that error term can be correlated over time within each state, but not across states. Further more , in all the regressions, heteroskedasticity robust standard errors has been used . V. EMPIRICAL RESULTS This section provides the main empirical results. In Table 1 the size of the government is proxied by the share of total state general expenditure in state personal income. Table 2 exploits an alternative measure of the size - the share of total public welfare expenditure in state personal income. TABLE 1 Dependent Variable expinc (1) (2) (3) (4) (5) (6) (7) (8) gip 0.0001565 (0.11) dem_houses -0.0030012 (-2.23)** -0.0018473 (-1.75)* -0.0033855 (-2.55)** 0.0003445 (0.37) 0.0007286 (0.54) gip x dem_houses -0.0004452 (-0.30) rep_houses -0.001542 (-0.74) divided_houses 0.0021876 (1.23) party_idx 0.0005012 (0.26) Time fixed effects NO YES NO YES YES YES YES YES State fixed effects NO NO YES YES YES YES YES YES Obs 1296 1296 1296 1296 1286 1296 1296 1286 R2 0.0039 0.0943 0.7883 0.8795 0.8802 0.8798 0.8799 0.8802 Adj R2 - 0.0750 0.7802 0.8722 0.8727 0.8725 0.8726 0.8729 *** Significant at the 1 % level ** Significant at the 5 % level * Significant at the 10 % level To start with the econometric analysis, column (1) provides results for simple OLS regression where the expenditure-to-income ratio is regressed simply on the dem_houses variable without controlling any additional factors. Ironically, the regression provides small but negative, statistically significant coefficient, which suggests that the leftist Democrats when in power are prone to lower government spending. However, it’s clear that the simple regression suffers from many omitted variables. and to deal with this bias, I employ time and state effects in columns 2 - 4. Noticeable that when both types of fixed effects are added to the regression, the coefficient turns positive, but it becomes to be statistically insignificant. Although, no causal relationship can be impugned from the regressions,what can be deemed is that a substantially higher R2 may suggest that introducing fixed effects increases the proportion of the variation in the dependent variable that can be explained by the explanatory variables. When there is a substantially high degree of correlation between explanatory variables,the use of several yet similar political variables that are available from my dataset into the regressions could bring about the problem of multicollinearity, Although this might not directly cause a problem for our OLS estimators, it raises some difficulty in separately identifying the effects and influences of particular variables. In column 5, I have accounted the effect of different branches of power – executive and legislative on government size. And the result of which is that the correlation between having control over the governor’s office and majority control over the legislative houses is fairly small1. Thus, it can be stipulated that the statistically insignificant interaction term gip x dem_houses suggests that there is no concrete or strong evidence which will directly point to the truism of the claim that a party’s control of both houses has bigger effect when also the governor’s office is also in the hands of the majority party. In relation to the level of government spending, the presence of rightist parties matters more compared with the presence of leftist parties. To substantiate this claim, I also regress the expenditure-to-income ratio on rep_houses that reflects dominance of the Republicans in both legislative houses. More over, in some cases divided government can substantially influence and affect fiscal outcomes for example through logrolling behavior. For this reason I create a variable divided_houses capturing situations when each party controls one legislative chamber. Likewise, I explore the relationship between the government size and the composite party index. However, all the three regressions provide insignificant coefficients. Summing up, despite using various specifications there is no substantial and robust evidence which points to the claim that a more left-oriented Democratic Party is more prone to increase in the total government expenditures relative to state income. TABLE 2 Dependent Variable welfinc welfexp (1) (2) (3) (4) (5) (6) (7) (8) gip 0.0000795 (0.16) 0.0045687 (1.74)* dem_houses 0.0018071 (2.56)** 0.0015987 (1.92)* 0.0119434 (2.02)** gip x dem_houses 0.0005407 (0.85) rep_houses -0.0015904 (-2.56)** -0.0099046 (-2.08)** divided_houses 0.0000055 (0.01) party_idx 0.0015948 (2.16)** 0.0138134 (2.65)*** Time fixed effects YES YES YES YES YES YES YES YES State fixed effects YES YES YES YES YES YES YES YES Obs 1296 1286 1296 1296 1286 1286 1296 1286 R2 0.8573 0.8561 0.8562 0.8536 0.8540 0.8520 0.8514 0.8514 Adj R2 0.8487 0.8471 0.8475 0.8447 0.8451 0.8428 0.8423 0.8423 *** Significant at the 1 % level ** Significant at the 5 % level * Significant at the 10 % level Similar pattern of empirical testing specification as in the previous table is used in Table 2, However, in this case the results suggest a modest impact of the leftist party in power on welfare expenditure-to-income ratio. The coefficient in column (1) suggest that when both legislative houses are in hand of the Democratic party the ratio is higher by 0.002, thus for every $1000 state income, the Democratic Party spend on welfare expenditure more by $2 than the Republican Party. The column (3) confirms the finding that the Republicans spend relatively less on welfare expenditures. The magnitude of this effect remains the same the when we use the estimation by the (Democratic) party composite index. The columns (6)-(8) shows that when the majority is being held by the leftist government, a higher governmental spending is allotted on welfare programs when compared with the rightist parties as the majority. This manifest concretely the divergence on the composition of expenditures, notably the welfare expenditure as a share of total state expenditure, is seen as one of the effects of political parties in government expenditure.. In addition to these findings, it safe to assume that the new government do not take a drastic shift in terms of expenditure at its preferred level. What it does undertake, on the other hand, is a reasonable shift in spending on projects or priorities it prefers. Furthermore, one factor that limits a drastic shift in spending is the working budget of the state which is basically based on the previous year budget. This point is being raised because of the supposition that the impact of political parties is technically perceptible in terms of welfare spending. that However, although the above statements are safely assumed , it should be noted that the researcher has not found any substantial improvement t in the robustness of results. VI ADDITIONAL EMPRICAL EVIDENCES The following table will show to us the distinct and mark differences in terms of political positions on important and significant issues between the Republican Party and Democratic Party. Table 1. Federal Government Spending ________________________________________________________________________ Federal Government Spending (in percent) Republicans Democrats The government in Washington should do everything possible to improve the standard of living of all Americans 47 76* Preferred size of federal government (rv):^ Smaller with fewer services 79* 43 Larger with many services 15 47* More important to you:^ Holding down size of government 50* 18 Providing needed services 43 76* Helping the middle class is very important in deciding 2000 presidential vote (rv)2 48 72* Paying doctor and hospital bills: Responsibility of government in Washington 33 65* People should take care of these themselves 31 * 8 Agree with both/neither 36* 27 What the federal government should do about increasing the number of Americans with health insurance (rv): Major effort that would require a tax increase 23 51 * Limited effort that would not require a tax increase 32 30 Not something the federal government should be doing 40* 11 Expanding health care coverage to all Americans is a moral issue (rv)^ 51 77* Sources: WP/Kaiser/Harvard 2000b. ^WP/Kaiser/Harvard 2000c. 3ABC/WP 2001 a. "Davis etal. 2000. =WP/Kaiser/Harvard 2000a. * Statistically significant difference (p < 0.05) rv = registered Source: Robert Blendon, John Benson, Mollyann Brodie, Drew Altman and Kathleen Weldon, ‘In Economic Policy How much does it matter which party governs?’, Challenge Nov.- Dec 2001. Under the issue of Federal Spending more Democrats (76%) are in favor of uplifting the quality of life of all Americans in all possible and that the burden or responsibility falls to Washington. While Republicans (46%) are more cautious in the ‘all possible means” as the method of achieving the good life for all Americans. In terms of government size the Republicans favor a smaller size and fewer services compared with the larger government size in providing services as supported by the Democrats. The table also presents to us that in terms of health care issues; three and quarters of Democrats hold it as a moral issue while half of the Republicans also see it in that way. TABLE 2. Government Regulation ________________________________________________________________________ Table 2 Government Regulation (in percent) Republicans Democrats Government has gone too far in regulating business (rv):^ Agree 82 49 Disagree 15 46* Government regulation does more harm than good^ 56 34 Government regulation protects the public^ 38 58 Government regulation of big business and corporations is necessary to protect the public^ 43 59 Corporations make reasonable profit^ 54 35 Corporations make too much profit^ 40 59 Immigration should be decreased" 51 38 Sources: WP/Kaiser/Harvard 2000c. 2Pew/PSRA 1999. ^WP/Kaiser/Harvard 1998. "Gallup Poll 2001a. * Statistically significant difference (p < 0.05) rv = registered voters _____________________________________________________________________________________ Source: Robert Blendon, John Benson, Mollyann Brodie, Drew Altman and Kathleen Weldon, ‘In Economic Policy How much does it matter which party governs?’, Challenge Nov.- Dec 2001. Under Government regulation, it appears that the moment the Democrats are the majority America can expected that there will be more regulations in the business world, especially in pharmaceutical, health, banking and insurance, as seen almost half of the Democrats believe that the government has not been regulating enough. It is also shown that for the Republicans 54% believes that corporations are gaining reasonable profits while 59% of Democrats believe that corporations are making too much profit TABLE 3 Economic Equity _____________________________________________________________________________ Table 3 Economic Equity (in percent) Republicans Democrats Responsibility of government to reduce income differences between those of high and low incomes: ^ 20 39* Income differences between rich and poor:^ Government in Washington ought to reduce differences 28 53* Government should not concern itself 57* 24 Are the following items moral issues (% rv answering yes) ?^ The differences in income between people with high incomes and people with low incomes 23 53 Taxes and the tax system 41 44 Money and wealth in this country (rv):" Distribution is fair 53 20 Should be more evenly distributed 43 75 Government should do more to help the needy^ 44 67 Government cant afford more help for the needy^ 48 27 Favor increasing minimum wage^ 74 93 It is the responsibility of the federal government to make sure minorities have jobs equal in quality to whites^ 32 59 Favor employers and colleges making an extra effort to find and recruit qualified minorities^ 48 64 Affirmative action;^ Race/ethnicity should be a factor when deciding who is hired, promoted, or admitted to college 2 8 Should be based strictly on merit 97 87 Because of past discrimination, blacks should be given preferences in hiring and promotion^ 27 11 It is the responsibility of the federal government to make sure blacks have schools equal in quality to whites^ 59 79* Sources: WP/Kaiser/Harvard 1998. ^Davis et al. 2000. ^WP/Kaiser/Harvard 2000b. "WP/Kaiser/Harvard 2000c. =Pew/PSRA1999. ••Pew/PSRA 2001 b. WP/Kaiser/Harvard 2001. * Statistically significant difference (p < 0.05) rv = registered voters 66 Challenge/November-December 2001 ______________________________________________________________________________________________ Source: Robert Blendon, John Benson, Mollyann Brodie, Drew Altman and Kathleen Weldon, ‘In Economic Policy How much does it matter which party governs?’, Challenge Nov.- Dec 2001. If we go over the economic equity it appears that for the Democrats ( 53%) think there is a disparity in the distribution of wealth in the America and that almost forty percent believes that the government should take a more active participation in the equitable distribution of wealth in the United States. But what is refreshing in this particular table that it appears that there is two issues of which both the Republican and the Democrats agree and that is one increase on minimum wage and the second is on the issue of affirmative action. Table 4 Global Economy and Trade Agreements "The New Economy" and Trade Agreements (in percent) Republicans Democrats Most jobs being created in the country today (rv):^ Pay well 41 39 Are low-paying 46 50 Globalization of world economy (rv): Is mostly good 38 36 Does not make much difference 20 31 Is mostly bad 29 17 Trade agreements between LInited States and other countries have (rv): Created U.S. jobs 22 20 Cost U.S. jobs 50 49 New economy really just helps people who are already rich (rv) 31 43 United States should have free trade with China on the same terms it gives its main trading partners:^ Agree 44 51 Disagree 49 43 Sources: WP/Kaiser/Harvard 2000c. ^ABC/WP 2001a. •Statistically significant difference (p < 0.05) rv = registered voters Source: Robert Blendon, John Benson, Mollyann Brodie, Drew Altman and Kathleen Weldon, ‘In Economic Policy How much does it matter which party governs?’, Challenge Nov.- Dec 2001. In this particular table it, it can be shown that both parties hold the belief that global trading and business is technically something that the government should pursue on the premises that it is beneficial on the American people as a whole. But what is notable in this particular table is the belief that although both a quarter of the republicans and Democrats believe that global economy has opened new opportunities for the Americans, almost half of both parties have opined that it has cost US jobs. VII FURTHER ELUCIDATION ON THE EMPRIRCAL EVIDENCES What we have seen in four tables that we have presented as proofs of the marked difference in positions between the Republican and the Democrats is the fact that although there are certain points in economic equity and on global trading and business it appears with clarity that both parties are on divergent positions on most pertinent issues that confounds contemporary America. And for this reason, there is a necessity on the part of political analysts and scholars to come up with a more viable explanation and solution on how the divided government of America can be resolved and ended. There is basically an urgent need to address this particular political situation on the premises that a divided or discordant political government is technically ineffectual in providing the necessary services that are mandated and needed in order to a better quality life to American people. Further more what the tables show is the truism that it is division of the government that is the primary factor in the decision and implementation of public policies In fact, the tables show that size of the government is one of the issues where the two big political parties in the US diverge. The real and authentic problem of America is its pluralistic multi-party system which have resulted into a divided government for US. Moreover, the four tables, though not encompassing in its scope in terms of the coverage of subject matter, what it highlights is the notion that although issues wherein both parties almost agrees there is a clear cut position with which the government may undertake. However, the moment that the issue starts to delve on economic matters wherein the two gigantic parties are technically in opposition, it appears that the only viable alternative is not non- recognition of the issue at hand but rather the tentative step that may be used as they try to come with a solution is a muddled compromise among the politicians themselves, with even the possibility of crossing party loyalties. VII. CONCLUSION In reading the article of Nouriel Roubini and Jeffrey Sachs entitled Government Spending and Budget Deficits in the Industrialised Countries it seems that they are raising the idea that multi-party system and government size are factors in policy making if the government is weak and divided, “weak and divided government ( as evidenced by the expected tenure in office, and by the number of political parties that share power in the governing coalition) have been less effective in reducing the budget deficit than stable and majority-party governments” (Roubini and Sachs, 1989, p 102). If we are going to correlate this with the divided government of United States, it seems that the case is not generally true. In effect what is being shown to us is that ‘weak and divided government’ is the most liable in committing errors in government spending and budgeting. The researcher is claiming this because the American government though it is a multi-party system is not a weak government for it can render services that are beneficial to its citizenry.. And this is attested by the fact that “many working Americans and their children have been living increasingly affluent lives. It is their prosperity that is so plainly visible in the ample houses of hundreds of suburbs across America in the proliferation of luxury vehicles and pleasure boats in palatial shopping centers and crowded holiday resorts”(Luttwak, 1999, p 3). What perhaps is the glaring reality is that as result of division in the government, there are certain delays in the delivery of the services certain tardiness in responses to urgent issues which have resulted to budget deficit (Alt, & Lowry, 1994) but it does not imply that division in the government is really detrimental to the government as a whole. In fact during 9/11 terror attack we have seen how political colors have been transcend as both the President and the Congress acted as one for the good of the nation as a whole. The terror attack is given as an instance wherein under difficult and extreme cases, the division in the American government is bridge. Thus, the researcher believes that in case of being weary and afraid of the possible repercussions of the division resulting from the multi-party system, it should be appreciated as the mark of the pluralistic society of which America draws its strength, stability and beauty. On the issue of governmental size, it has been shown that basically in the United States government size depends on who occupies the highest political position. Government size becomes a problem if it is related to budget. But in the case of United States, their bigger problem is on how they can resolve the government divide since if there is no government divide then size of the government is more or less stabilised. This is being claimed on the supposition that we know that the Republicans prefer a smaller government whereas the Democrats prefer a larger one. So, whoever sits in Washington determines the government size. And this particular scenario is what the paper is driving at. Multi-party system becomes a very big issue the moment the political parties come to a stalemate. Division in government becomes apparent when there is a crisis where political parties take a divergent position. As such, one of the major hurdles of the American government is the division, the multi-party system that it applies in its political exercise. And that the size of the government is a problem as a result of the divide but it is not to be seen as a separate, distinct problem from the problem of division in the government. More over, the results suggest that the presence of leftist political parties in the government does not have a significant effect on total state spending relative to the rightist parties. Thus, my findings does not fully support the results of Roubini et al. (1989). Nevertheless, this study finds a modest effect of parties’ political ideology on welfare spending relative to state total income. United States government indeed is divided but then again it must not be forgotten that the contemporary world thrives in plurality and that the political set up of United States has consistently show the world of the actuality of attaining the good life in the midst of diversity, in the presence of the division, the difference. REFERENCES: Arendt, H.(1958) The Human Condition. Chicago: University of Chicago Press. Alt, J. & Lowry, R.C. (1994) ‘Divided government fiscal institutions and budget deficits: Evidence from the States’,The American Political Science Review, Vol 88, No 4, p811. Brady, D.W. ( 1993) ‘ The causes and consequences of divided government: Toward a new theory of American politics’, The American Political Science Review, Vol 87 No 1. Blendon, R., Benson, J., Brodie, M., Altman, D., & Weldon, K. (2001) ‘ In economic policyhow much does it matter which party governs?’, Challenge, Vol 44, No 6, p 59 – 77. Feldstein, M. (1992) ‘The council of economic advisers and economic advising in the United States’, The Economic Journal, 102, pp 1223- 1234. Luttwak E. (1999). The hidden effects of free-market capitalism: Turbo capitalism, winners and losers in the global economy. New York: HarperCollins. Mixon, F. Jr & Hobson, D.L. (2001) ‘Intergovernmental grants and the positioning of presidential primaries and caucuses: Empirical evidence from the 1992, 1996, and 2000 election cycles’, Contemporary Economic Policy, Vol 19 No 1, p 27-38. Roubini, N., & Sachs, J. (1989) ‘Government spending and budget deficits in the industrial countries’, Economic Policy. Wright, G. (1973) ’The political economy of new deal spending: An econometric analysis’, The Review of Economics and Statistics. Electronic Sources: List of Political parties in the United States. Retrieved on March 12, 2007 from http://www.en.wiipedia.org/wiki/List_of_political_ parties_in_the_United_States. Read More