Welcome! I am an Assistant Professor in the Department of Political Science at the University of Colorado Boulder. I received my Ph.D in Political Science from Texas A&M University in 2017.
I specialize in political economy and methodology. My current research interests focus on the manipulation of budgets during elections, approaches to modeling compositional data (e.g., budgets or party vote support), and developing and implementing models for time series and cross-sectional time series data.
On this website you can find my CV, links to my work, software, a variety of downloads for both Stata and R, and course materials. You can also contact me or check out my Google Scholar page.
Philips, Andrew Q. 2020. "An easy way to create duration variables in binary cross-sectional time series data." The Stata Journal 20(4): 916-930.
In cross-sectional time-series data with a dichotomous dependent variable, failing to account for duration dependence when it exists can lead to faulty inferences. A common solution is to include duration dummies, polynomials, or splines to proxy for duration dependence. Because creating these is not easy for the common practitioner, I introduce a new command, mkduration, that is a straightforward way to generate a duration variable for binary cross-sectional time-series data in Stata. mkduration can handle various forms of missing data and allows the duration variable to easily be turned into common parametric and nonparametric approximations.
Jordan, Soren and Andrew Q. Philips. 2020. "Exploring meaningful visual effects and quantities of interest from dynamic models through dynamac." The Journal of Open Source Software 5(54): 2528.
dynamac (2020) implements a framework of estimating and interpreting autoregressive distributed lag (ARDL) models in R. dynamac uses stochastic simulation techniques (Jordan & Philips 2018a,2018b) to easily recover traditional quantities of interest, such as short- and long-run effects, even from complex dynamic specifications, including models with cointegration. These simulation techniques bring a wider set of inferences from complex models to users in fields as diverse as environmental science (Danish 2020) and economics (Sharma 2019). Users can make dynamic inferences from their models, including measures of uncertainty, without the need for any formulae or algebraic solutions.
Jung, Yoo Sun, Flávio D.S. Souza, Andrew Q. Philips, Amanda Rutherford and Guy D. Whitten. 2020. "A program to estimate and interpret models of dynamic compositional dependent variables: New features for dynsimpie" The Stata Journal 20(3): 584-603.
Philips, Rutherford, and Whitten (2016, Stata Journal 16: 662–677) introduced dynsimpie, a command to examine dynamic compositional dependent variables. In this article, we present an update to dynsimpie and three new adofiles: cfbplot, effectsplot, and dynsimpiecoef. These updates greatly enhance the range of models that can be estimated and the ways in which model results can now be presented. The command dynsimpie has been updated so that users can obtain both prediction plots and change-from-baseline plots using postestimation commands. With the new command dynsimpiecoef, various types of coefficient plots can also be obtained. We illustrate these improvements using monthly data on support for political parties in the United Kingdom.
Philips, Andrew Q. 2020. "Just in time: Political policy cycles of land reform." Politics 40(2): 207-226.
While the political budget cycle literature focuses on the manipulation of existing policies, an analysis of the impact of the passage of redistributive policies themselves remains absent. I contend that policy passage is a strategically timed signal to voters used before elections to benefit the incumbent. Using aggregate data on land reforms in India from 1957 to 1992, I find that reforms are indeed timed before elections. Second, using historical survey data, I show that land issues remain a strong signal to Indian voters over time, even in states that have already enacted reforms. These findings provide evidence for political policy cycles.
Benton, Allyson L., and Andrew Q. Philips. 2020. "Does the @realDonaldTrump really matter to financial markets?" American Journal of Political Science 64(1): 169-190.
Does the @realDonaldTrump really matter to financial markets? Research shows that new information about the likely future policy direction of government affects financial markets. In contrast, we argue that new information can also arise about the likely future government's resolve in following through with its policy goals, affecting financial markets as well. We test our argument using data on U.S. President Donald J. Trump's Mexico‐related policy tweets and the U.S. dollar/Mexican peso exchange rate. We find that Trump's Mexico‐related tweets raised Mexican peso volatility while his policy views were unknown as well as thereafter, as they signaled his resolve in carrying out his Mexico‐related agenda. By helping politicians disseminate policy information to voters, and since voters hold governments accountable for their policy performance, social media allows investors to gather information about the likely policy direction and policy resolve of government, especially those of newcomers whose direction and resolve are unknown.
Philips, Andrew Q., Flavio D.S. Souza, and Guy D. Whitten. 2020. "Globalization and comparative compositional inequality." Political Science Research and Methods 8(3): 509-525.
Globalization has been one of the biggest driving forces of the last half century. There has been substantial disagreement about the impact that increased international integration has on income inequality. Though most agree that globalization positively affects economic output, it is no surprise that it leads to relative winners and losers within nations. The question that remains is where in the income distribution are these relative gains and losses occurring? We offer a broader picture of globalization’s effects on inequality by using a dynamic compositional approach to test the impact of globalization and relative factor endowments on the composition of income. Using data from four countries, we model the effects of globalization on quantiles of the income distribution. Our findings suggest that globalization has substantial (and divergent) effects across income strata, and that these effects differ across nations based on relative factor endowments.
Lipsmeyer, Christine S., Andrew Q. Philips, Amanda Rutherford, and Guy D. Whitten. 2019. "Comparing dynamic pies: A strategy for modeling compositional variables in time and space." Political Science Research and Methods 7(3): 523-540.
Across a broad range of fields in political science, there are many theoretically interesting dependent variables that can be characterized as compositions. We build on recent work that has developed strategies for modeling variation in such variables over time by extending them to models of time series cross-sectional data. We discuss how researchers can incorporate the influence of contextual variables and spatial relationships into such models. To demonstrate the utility of our proposed strategies, we present a methodological illustration using an analysis of budgetary expenditures in the US states.
Funk, Kendall D. and Andrew Q. Philips. 2019. "Gendered budgeting: Women chief executives and budget allocations in local governments." Political Research Quarterly 72(1): 19-33.
One potential consequence of increasing women's numeric representation is that women elected officials will behave differently than their men counterparts and improve women's substantive representation. This study examines whether electing women to local offices changes how local government expenditures are allocated in ways that benefit women. Using compositional expenditure data from more than 5,400 Brazilian municipalities over eight years, we find significant differences in the ways men and women mayors allocate government expenditures. Our findings indicate that women mayors spend more on traditionally feminine issues, and less on traditionally masculine issues, relative to men mayors. In regard to specific policy areas, we find that women spend more on women's issues, including education, health care, and social assistance, and less on masculine issues, including transportation and urban development, relative to men mayors. We further find that women's legislative representation significantly influences the allocation of expenditures as a larger percentage of women councilors increases spending on traditionally feminine issues, as well as education, health care, and social assistance, relative to other policy issues. These findings indicate that women local elected officials improve women's substantive representation by allocating a larger percentage of expenditures to issues that have historically and continue to concern women in Brazil.
Jordan, Soren and Andrew Q. Philips. 2018. "Dynamic simulation and testing for single-equation cointegrating and stationary autoregressive distributive lag models." The R Journal 10(2): 469-488.
While autoregressive distributed lag models allow for extremely flexible dynamics, interpret ing the substantive significance of complex lag structures remains difficult. In this paper we discuss dynamac (dynamic autoregressive and cointegrating models), an R package designed to assist users in estimating, dynamically simulating, and plotting the results of a variety of autoregressive distributed lag models. It also contains a number of post-estimation diagnostics, including a test for cointegration for when researchers are estimating the error-correction variant of the autoregressive distributed lag model.
Jordan, Soren and Andrew Q. Philips. 2018. "Cointegration testing and dynamic simulations of autoregressive distributed lag models." Stata Journal 18(4): 902-923.
In this article, we introduce dynamac, a suite of commands designed to assist users in modeling and visualizing the effects of autoregressive distributed lag models and in testing for cointegration. We discuss the bounds cointegration test proposed by Pesaran, Shin, and Smith (2001, Journal of Applied Econometrics 16: 289–326), which we have adapted into a command. Because the resulting models can be dynamically complex, we follow the advice of Philips (2018, American Journal of Political Science 62: 230–244) by introducing a flexible command designed to dynamically simulate and plot a variety of types of autoregressive distributed lag models, including error-correction models.
Philips, Andrew Q. 2018. "Have your cake and eat it too? Cointegration and dynamic inference from autoregressive distributed lag models." American Journal of Political Science 62(1): 230-244.
Although recent articles have stressed the importance of testing for unit-roots and cointegration in time series analysis, practitioners have been left without a straightforward procedure to implement this advice. I propose using the autoregressive distributed lag model and bounds cointegration test as an approach to dealing with some the most commonly encountered issues in time series analysis. Through Monte Carlo experiments I show that this procedure performs better than existing cointegration tests under a variety of situations. I illustrate how to implement this strategy with two step-by-step replication examples. To further aid users, I have designed software programs in order to test and dynamically model the results from this approach.
Lipsmeyer, Christine S., Andrew Q. Philips, and Guy D. Whitten. 2017. "The effects of immigration and integration on European budgetary trade-offs." Journal of European Public Policy 24(6): 912-930.
What affects government policy-making continues to be an important question for researchers interested in political competition and policy priorities. In this contribution, we bring together a theoretical framework that focuses on the influence of globalizing forces on government policy decisions with a methodological emphasis on explaining dynamic budgetary trade-offs. While comparative public policy and budgetary scholars typically have focused on entire budgets or amounts spent on specific policies, we look at the political priorities embedded in budgets by modeling the budget as a pie. Then, we theorize about how governments respond to external shocks by altering the allocations to the various policy areas. Using this approach, we find that governments of different ideological types react to external shocks by altering their different policy priorities.
Philips, Andrew Q. 2016. "Seeing the forest through the trees: A meta-analysis of political budget cycles." Public Choice 168(3) 313-341.
Despite a vast number of articles, the political budget cycle literature contains many conflicting theories and empirical results. I conduct the first ever meta-analysis of this literature in order to establish whether a link between elections and government budgets exists. Using data on 1198 estimates across 88 studies published between 2000 and 2015, I find evidence of a statistically significant---yet substantively small---increase in government expenditures and public debt around elections, and reductions in revenues and fiscal balance. Using meta-regression analysis combined with Bayesian model averaging, I find support for some of the context-conditional theories in the literature. Although the findings of political budget cycles are robust to publication bias as well as some of the methodological- and study-specific choices authors are forced to make, they also shed light on how certain decisions may affect a study's findings. This has implications for current and future research on political budget cycles.
Philips, Andrew Q., Amanda Rutherford, and Guy D. Whitten. 2016. "Dynamic pie: A strategy for modeling trade-offs in compositional variables over time." American Journal of Political Science 60(1): 268-283.
The substance of politics involves competition that evolves over time. While our theories about competition emphasize trade-offs across multiple categories, most empirical models tend to oversimplify them by considering trade-offs between one category and everything else. We propose a research strategy for testing theories about trade-off relationships that shape dynamic compositional variables. This approach improves current methods used to analyze compositional dependent variables by addressing two limitations. First, although scholars have considered compositional dependent variables, they have done so in contexts that were not dynamic. Second, current approaches toward graphical presentations become unwieldy when the compositional dependent variable has more than three categories. We demonstrate the utility of our strategy to expand current theories of party support and political budgeting. In both cases, we can extend trade-offs across pairs of alternatives (e.g., prime minister versus all other parties or spending on defense versus everything else) to competition across multiple alternatives.
Philips, Andrew Q., Amanda Rutherford, and Guy D. Whitten. 2015. "The dynamic battle for pieces of pie: Modeling party support in multi-party nations." Electoral Studies 39: 264-274.
When teams of rival politicians compete for public support, they are essentially playing a zero sum game where one party's gains tend to come from the losses of one or more of their opponents. Despite this, most analyses of party support across time model the dynamics associated with a single party's support. In nations where only two parties are competing for votes, this approach is fine. But in nations with more than two parties, much of the substance of what is going on in party competition is lost. In this paper we illustrate the usefulness of a modeling strategy proposed by Philips et al. (2015) for estimating and interpreting the causal relationships that shape trade-offs in party support as they evolve over time. We extend their work by modeling public support for four parties instead of three and by developing the ability to model dynamic changes in party characteristics. We estimate our models on monthly data from the United Kingdom and Germany.