This thesis consists of a summary and four papers. The first two concerns health care and sickness absence, and the last two pharmaceutical costs and prices. Paper [I] presents an economic federation model which resembles the situation in, for example, Sweden. In the model the state governments provide health care, the fed-eral government provides a sickness benefit and both levels tax labor income. The re-sults show that the states can have either an incentive to under- or over-provide health care. The federal government can, by introducing an intergovernmental transfer, in-duce the state governments to provide the socially optimal amount of health care. In Paper [II] the effect of aggregated public health care expenditure on absence from work due to sickness or disability was estimated. The analysis was based on data from a panel of the Swedish municipalities for the period 1993-2004. Public health care expenditure was found to have no statistically significant effect on absence and the standard errors were small enough to rule out all but a minimal effect. The result held when separate estimations were conducted for women and men, and for absence due to sickness and disability. The purpose of Paper [III] was to study the effects of the introduction of fixed pharmaceutical budgets for two health centers in Västerbotten, Sweden. Estimation results using propensity score matching methods show that there are no systematic differences for either price or quantity per prescription between health centers using fixed and open-ended budgets. The analysis was based on individual prescription data from the two health centers and a control group both before and after the introduction of fixed budgets. In Paper [IV] the introduction of the Swedish substitution reform in October 2002 was used as a natural experiment to examine the effects of increased consumer infor-mation on pharmaceutical prices. Using monthly data on individual pharmaceutical prices, the average reduction of prices due to the reform was estimated to four percent for both brand name and generic pharmaceuticals during the first four years after the reform. The results also show that the price adjustment was not instant.

El proceso de apertura comercial experimentado por Argentina en los años 90 sometió a la industria por primera vez en mucho tiempo a la competencia externa. Este trabajo investiga en qué medida la competencia externa potencial actuó como un limitante del poder de mercado de la industria local. Utilizando modelos dinámicos para datos en panel se demostró que dicha competencia fue significativa durante el período 1995-2001. En particular, algunos de los modelos estimados sugieren que aquellas industrias menos concentradas, con menor protección arancelaria, con economías de escala más reducidas, no vinculadas directamente a la actividad agropecuaria o que producen productos menos diferenciados son las que enfrentaron una competencia potencial externa significativa. / The process of trade liberalization experienced by Argentina in the 90s subjected domestic industries to foreign competition for the first time in many years. This paper studies to what extent potential foreign competition was a limiting factor of domestic market power. Using dynamic panel data models, this paper shows that such competition was significant between 1995 and 2001. Specifically, some of the models suggest that those industries not directly linked to agricultural activities, with a low degree of concentration, with low tariff barriers, with a low degree of economies of scale and whose output has a low degree of differentiation are the ones that experienced significant potential foreign competition. / Una versión de este trabajo obtuvo el premio "Elías Salama" otorgado en las X Jornadas de Economía Monetaria e Internacional, organizadas por el Departamento de Economía de la Universidad Nacional de La Plata.

Made available in DSpace on 2010-04-20T21:00:45Z (GMT). No. of bitstreams: 4 Reginaldo Togashi Takara.pdf.jpg: 2588 bytes, checksum: d3064a4acbf672d14b9e5acaf7dc7adb (MD5) Reginaldo Togashi Takara.pdf.txt: 361380 bytes, checksum: f20271e10bed2dc4d7f1e87e463b580e (MD5) license.txt: 4712 bytes, checksum: 4dea6f7333914d9740702a2deb2db217 (MD5) Reginaldo Togashi Takara.pdf: 767400 bytes, checksum: 75fce885020db2ada0e370808d1d01df (MD5) Previous issue date: 2010-02-05T00:00:00Z / This dissertation studies the relationship between economic growth, natural resources and export diversification. We build four variables of concentration for total exports, natural resource exports and renewable and nonrenewable resource exports using a panel of 93 countries over 1965-2005. Then we estimate a dynamic panel model using a two-step system GMM approach (Sys-GMM2). The results indicate a significant and positive relationship between natural resource export diversity and economic growth, while the share of natural resources in gross national product (GNP) is not statistically significant. This result suggests that resource-abundant countries can, in principle, promote growth by diversifying exports into new primary or resource-related products, even when natural resources account for a significant share of GNP, thus avoiding the so-called “natural resource curse”. Our conclusions come in line with recent theoretical and empiric studies on both natural resources and export diversification. / Esta dissertação estuda a relação entre crescimento econômico, recursos naturais e diversificação das exportações. Para tanto, são construídas quatro variáveis que medem a concentração das exportações totais, das exportações de recursos naturais e das de recursos renováveis e não-renováveis para um painel de 93 países cobrindo o período 1965-2005. Em seguida, um modelo de painel dinâmico é estimado pela abordagem GMM em sistema e em dois estágios (Sys-GMM2). Os resultados indicam que há uma relação positiva e significante entre diversidade das exportações de recursos naturais e crescimento econômico, enquanto que a participação de recursos naturais no produto nacional bruto (PNB) não é estatisticamente significante. Esse resultado sugere que países com abundância de recursos naturais podem, a princípio, promover o crescimento diversificando suas exportações com novos produtos primários ou em setores relacionados, mesmo quando os recursos naturais representem parcela significativa do PNB, evitando assim a chamada "maldição dos recursos naturais". As conclusões se alinham com estudos teóricos e empíricos recentes sobre recursos naturais e diversificação das exportações.

As conclusões das pesquisas sobre velocidade de ajustamento da estrutura de capital em direção ao alvo (SOA - Speed Of Adjustment) são muito divergentes quanto à velocidade medida. A maioria dos estudos mensura o SOA sem levar em conta certas especificidades das empresas ou de suas estratégias. Dividindo-se a amostra em empresas não adquirentes, adquirentes eventuais e adquirentes em série, o presente estudo tem por objetivo investigar o padrão de comportamento do SOA no período entre 1990 e 2010 para empresas norte-americanas. Considerando diversas variáveis de controle (restrições financeiras, oportunidades de crescimento, grau de alavancagem financeira e emissão/redução de dívidas e ações), tem-se uma investigação mais aprofundada de como as empresas não adquirentes, adquirentes eventuais e adquirentes em série ajustam sua estrutura de capital em direção ao alvo. São utilizadas como variáveis dependentes o endividamento a valor de mercado e endividamento líquido a valor de mercado. Verificou-se que os endividamentos médios para as três amostras são diferentes entre si, de forma estatisticamente significante. As empresas adquirentes em série possuem o menor endividamento, ao passo que as empresas não adquirentes são as mais endividadas, estando as empresas adquirentes eventuais em posição intermediária quanto ao endividamento. Este resultado sugere que as empresas possuem padrões de endividamentos distintos em função de sua política de investimentos relacionada à aquisições. É utilizado o modelo de ajustamento parcial para mensurar o SOA, empregando painel de dados dinâmico com a técnica do Método dos Momentos Generalizados (GMM) Sistêmico para se medir a velocidade de ajustamento da estrutura de capital de um período a outro. Esta técnica tem-se mostrado a menos enviesada e, assim, tem sido uma das mais utilizadas em trabalhos empíricos. Os resultados encontrados evidenciam que o padrão de comportamento do SOA pode depender da sua estratégia de investimentos em aquisições. O SOA das empresas adquirentes em série é constantemente menor em comparação ao SOA das empresas adquirentes, mesmo considerando as diversas variáveis de controle. O SOA das empresas não adquirentes permanece em posição intermediária. Esses resultados em conjunto sugerem que a folga financeira (baixo nível de endividamento e elevado saldo de caixa) seja um fator relevante para as adquirentes em série. Desse modo, tais empresas ajustam seu endividamento de forma mais lenta, em resposta a uma estrutura de capital mais adequada à sua política de investimentos. Por outro lado, as empresas adquirentes eventuais ajustam mais rapidamente sua estrutura de capital em função de sua política de aquisições esporádicas. / The conclusions of surveys on speed of adjustment (SOA) towards the target capital structure are widely divergent as regards the measured speed. Most studies measure the SOA without taking into account certain specificities of the companies or of their strategies. Dividing the sample in non-acquiring companies, sporadic acquiring companies and serial acquirers, this study seeks to investigate the pattern of behavior of the SOA, in the period between 1990 and 2010 for North American companies. Considering several control variables (financial restrictions, growth opportunities, degree of financial leverage and issuance/reduction of debts and shares), we have a more in-depth investigation into how non-acquiring companies, sporadic acquiring companies and serial acquirers adjust their capital structure towards the target. Market leverage value and net market leverage are used as dependent variables. It was verified that the average debt levels for the three samples are different from one another, in a statistically significant manner. The serial acquirers have the lowest leverage, while non-acquiring companies have the highest leverage, with acquirers companies in an intermediate position on the leverage. This result suggests that companies have different debts patterns, due to its investment policy related to acquisitions. The partial adjustment model is employed to measure the SOA, using dynamic panel data with the Generalized Method of Moments (GMM) Systemic to measure the speed of adjustment of the capital structure from one period to another. This technique has shown itself to be less biased and has thus been one of the most commonly used techniques in empirical studies. The results show evidence that the SOA may depend of its investment strategy. The SOA of serial acquirers companies is constantly lower than the SOA of acquiring companies, even considering all the control variables. The SOA of non-acquiring companies remains in an intermediate position. Taken together, these results suggest that the financial slack (low debt and high cash balance) is a relevant factor to serial acquirers. Thus such companies adjust their debt more slowly in response to a more adequate capital structure to their investment policy. On the other hand, sporadic acquiring companies adjust faster its capital structure, due to its sporadic acquisition policy.

The constant increase in public sector spending in the advanced economies, increases pressure on the revenue side of public budgets. The primary sources of public budgets are taxes. This raises a question of what type of tax instrument to choose in respect to meet the high efficiency in sourcing public budgets restriction and at the same time, to minimize the negative impact on the private sector and households. As generally effective tax is considered a consumption tax. But what is the real effect of this tax in the short run and long run? The aim of this thesis is to analyze the influence of value -- added tax (VAT) on household expendictures in the short and long run. At first I provide empirical test of the short-term effect of VAT on household spending using quarter panel data for Visegrad fore countries, as I am focused mostly on the Czech Republic and Slovakia. Then construct a broader set of data for 14 EU countries, in purpose to test the long-term effect. For testing, I use several estimation techniques for panel data, taking into account the dynamic nature of these data sets.

This thesis examines the effectiveness of macroprudential policies on reducing housing price growth in the international database of 56 countries with the use of GMM and fixed effects between 2000 and 2017. The macroprudential index is added to the dynamic panel data model where the housing price index is regressed on housing price determinants as the economic growth or unemployment rate. The analysis is also conducted on the sample of countries with a higher market share of owners with a mortgage as there is a higher opportunity to control the housing market through the credit channel. Nevertheless, results show that we do not have enough evidence to state that macroprudential policies curb house price booms. Contrarily, the effect seems to work in the opposite direction which is probably caused by a reverse causality between the growth of real estate prices and the implementation of macroprudential tools. The debt-to-income restriction is the only tool that decreases housing price growth according to the fixed effects model. Detailed counterfactual analysis of the Czech market proposes only a slight impact of the loan-to-value measure on the apartment price development according to one out of four predictions. 1

The study aims to investigate the impact of natural resource wealth on the shadow economy. The theoretical section provides the basis of understanding the nature of two phenomena and discusses the possible transmission channels through which natural resources might influence the shadow economy. Consequently, the key determinants of the shadow economy are examined by static and dynamic models. Natural resource abundance is proxied by natural resource rents. We employ a panel data set for 109 countries for the period from 1996 to 2006. The results reveal that resource wealth is associated with the decrease of the shadow economy. This result is robust for different resource types (durable and non-durable), and the effect is more profound for countries with a low income level. JEL Classification C33, E26, O13 Keywords natural resources, shadow economy, dynamic panel data models, system GMM estimator Title Natural Resource Curse and Shadow Economy: Empirical Evidence

本篇論文在結構性時間序列模型基礎下，將中國大陸29省市自治區1978-2005年實質人均GDP，拆解出其長期趨勢變動軌跡中的水準值與斜率值，對照傳統上直接利用實質所得數據，以動態縱橫資料方法進行經濟成長條件收斂假說的檢定。本文特色在於加入潛在GDP長期趨勢項的水準值和斜率值，並利用內生解釋變數落後項動態分析。除可驗證隨著時間經過，中國相對貧窮省區是否終將逐漸趕上相對富有省份所得水準外，其次，根據GDP趨勢項一階與二階條件的收斂與否，可進而確認實質GDP收斂的本質。 我們發現，實質人均GDP收斂的本質關鍵在於潛在趨勢水準收斂，潛在GDP趨勢斜率的成長率將左右區域間實質所得收斂速度。大部分樣本中，擴大的Solow模型或考慮不同經濟開放程度因素下的內生成長模型，支持條件收斂假說，而後者設算出的收斂係數明顯較為低。此外，考慮採用Arellano and Bond(1991)的the first difference GMM估計式可能存在弱工具性問題(a weak instruments problem)，以Blundell and Bond(1998)發展出的the system GMM估計式，作為探討初始所得與經濟成長收斂的關係應是較為適合的方法。 / This research examines the economic growth conditional convergence hypothesis. Using the data of 29 provinces in Mainland China between 1978 and 2005, this study applied the structural time series model to deconstruct the provinces’ real GDP per capita into two parts - the level and the slope of trend movement. The characteristics of this paper are to include the level and the slope of trend of potential GDP and to consider the lagged dependent variables into the panel data. This study intends to validate whether the income level of relatively poor provinces will gradually catch up that of the relatively affluent provinces in Mainland China eventually. In addition, this study, based on the convergence or divergence in the first-order and second-order conditions of GDP tendency, will confirm the essence of the convergence in real GDP. The findings are that the essential key of the convergence in real GDP per capita is the convergence of the potential level of GDP. The growth of potential GDP tendency slope would affect the converging speed of real income in regions. The testing results of either the augmented Solow model or the endogenous growth model which considered different economic opening degrees both support the conditional convergence hypothesis in most sample sets, while the estimated convergence coefficients of the later are significantly lower than those of the former. In addition, considering the possible weak instruments problem in the first difference GMM estimator developed by Arellano and Bond (1991), the system GMM developed by Blundell and Bond (1998) should be a more suitable way to observe the relation between initial income level and economic growth convergence.

This thesis consists of three chapters on economic and econometric applications of Bayesian parameter estimation and model comparison. The first two chapters study the incidental parameter problem mainly under a linear autoregressive (AR) panel data model with fixed effect. The first chapter investigates the problem from a model comparison perspective. The major finding in the first chapter is that consistency in parameter estimation and model selection are interrelated. The reparameterization of the fixed effect parameter proposed by Lancaster (2002) may not provide a valid solution to the incidental parameter problem if the wrong set of exogenous regressors are included. To estimate the model consistently and to measure its goodness of fit, the Bayes factor is found to be more preferable for model comparson than the Bayesian information criterion based on the biased maximum likelihood estimates. When the model uncertainty is substantial, Bayesian model averaging is recommended. The method is applied to study the relationship between financial development and economic growth. The second chapter proposes a correction function approach to solve the incidental parameter problem. It is discovered that the correction function exists for the linear AR panel model of order p when the model is stationary with strictly exogenous regressors. MCMC algorithms are developed for parameter estimation and to calculate the Bayes factor for model comparison. The last chapter studies how stock return's predictability and model uncertainty affect a rational buy-and-hold investor's decision to allocate her wealth for different lengths of investment horizons in the UK market. The FTSE All-Share Index is treated as the risky asset, and the UK Treasury bill as the riskless asset in forming the investor's portfolio. Bayesian methods are employed to identify the most powerful predictors by accounting for model uncertainty. It is found that though stock return predictability is weak, it can still affect the investor's optimal portfolio decisions over different investment horizons.

Ma thèse est composée de trois essais sur l'inférence par le bootstrap à la fois dans les modèles de données de panel et les modèles à grands nombres de variables instrumentales #VI# dont un grand nombre peut être faible. La théorie asymptotique n'étant pas toujours une bonne approximation de la distribution d'échantillonnage des estimateurs et statistiques de tests, je considère le bootstrap comme une alternative. Ces essais tentent d'étudier la validité asymptotique des procédures bootstrap existantes et quand invalides, proposent de nouvelles méthodes bootstrap valides. Le premier chapitre #co-écrit avec Sílvia Gonçalves# étudie la validité du bootstrap pour l'inférence dans un modèle de panel de données linéaire, dynamique et stationnaire à effets fixes. Nous considérons trois méthodes bootstrap: le recursive-design bootstrap, le fixed-design bootstrap et le pairs bootstrap. Ces méthodes sont des généralisations naturelles au contexte des panels des méthodes bootstrap considérées par Gonçalves et Kilian #2004# dans les modèles autorégressifs en séries temporelles. Nous montrons que l'estimateur MCO obtenu par le recursive-design bootstrap contient un terme intégré qui imite le biais de l'estimateur original. Ceci est en contraste avec le fixed-design bootstrap et le pairs bootstrap dont les distributions sont incorrectement centrées à zéro. Cependant, le recursive-design bootstrap et le pairs bootstrap sont asymptotiquement valides quand ils sont appliqués à l'estimateur corrigé du biais, contrairement au fixed-design bootstrap. Dans les simulations, le recursive-design bootstrap est la méthode qui produit les meilleurs résultats. Le deuxième chapitre étend les résultats du pairs bootstrap aux modèles de panel non linéaires dynamiques avec des effets fixes. Ces modèles sont souvent estimés par l'estimateur du maximum de vraisemblance #EMV# qui souffre également d'un biais. Récemment, Dhaene et Johmans #2014# ont proposé la méthode d'estimation split-jackknife. Bien que ces estimateurs ont des approximations asymptotiques normales centrées sur le vrai paramètre, de sérieuses distorsions demeurent à échantillons finis. Dhaene et Johmans #2014# ont proposé le pairs bootstrap comme alternative dans ce contexte sans aucune justification théorique. Pour combler cette lacune, je montre que cette méthode est asymptotiquement valide lorsqu'elle est utilisée pour estimer la distribution de l'estimateur split-jackknife bien qu'incapable d'estimer la distribution de l'EMV. Des simulations Monte Carlo montrent que les intervalles de confiance bootstrap basés sur l'estimateur split-jackknife aident grandement à réduire les distorsions liées à l'approximation normale en échantillons finis. En outre, j'applique cette méthode bootstrap à un modèle de participation des femmes au marché du travail pour construire des intervalles de confiance valides. Dans le dernier chapitre #co-écrit avec Wenjie Wang#, nous étudions la validité asymptotique des procédures bootstrap pour les modèles à grands nombres de variables instrumentales #VI# dont un grand nombre peu être faible. Nous montrons analytiquement qu'un bootstrap standard basé sur les résidus et le bootstrap restreint et efficace #RE# de Davidson et MacKinnon #2008, 2010, 2014# ne peuvent pas estimer la distribution limite de l'estimateur du maximum de vraisemblance à information limitée #EMVIL#. La raison principale est qu'ils ne parviennent pas à bien imiter le paramètre qui caractérise l'intensité de l'identification dans l'échantillon. Par conséquent, nous proposons une méthode bootstrap modifiée qui estime de facon convergente cette distribution limite. Nos simulations montrent que la méthode bootstrap modifiée réduit considérablement les distorsions des tests asymptotiques de type Wald #$t$# dans les échantillons finis, en particulier lorsque le degré d'endogénéité est élevé. / My dissertation consists of three essays on bootstrap inference in both large panel data models and instrumental variable (IV) models with many instruments and possibly, many weak instruments. Since the asymptotic theory is often not a good approximation to the sampling distribution of test statistics and estimators, I consider the bootstrap as an alternative. These essays try to study the asymptotic validity of existing bootstrap procedures and when they are invalid, to propose new valid bootstrap methods. The first chapter (co-authored with Sílvia Gonçalves) studies the validity of the bootstrap for inference on a stationary linear dynamic panel data model with individual fixed effects. We consider three bootstrap methods: the recursive-design wild bootstrap, the fixed-design wild bootstrap and the pairs bootstrap. These methods are natural generalizations to the panel context of the bootstrap methods considered by \citeasnoun{GK} in pure time series autoregressive models. We show that the recursive-design wild bootstrap fixed effects OLS estimator contains a built-in bias correction term that mimics the incidental parameter bias. This is in contrast with the fixed-design wild bootstrap and the pairs bootstrap whose distributions are incorrectly centered at zero. As it turns out, both the recursive-design and the pairs bootstrap are asymptotically valid when applied to the bias-corrected estimator, but the fixed-design bootstrap is not. In the simulations, the recursive-design bootstrap is the method that does best overall. The second chapter extends our pairwise bootstrap results to dynamic nonlinear panel data models with fixed effects. These models are often estimated with the Maximum Likelihood Estimator (MLE) which also suffers from an incidental parameter bias. Recently, \citeasnoun{DhaeneJochmans} have proposed the split-jackknife estimation method. Although these estimators have asymptotic normal approximations that are centered at the true parameter, important size distortions remain in finite samples. \citeasnoun{DhaeneJochmans} have proposed the pairs bootstrap as an alternative in this context without a theoretical justification. To fill this gap, I show that this method is asymptotically valid when used to estimate the distribution of the half-panel jackknife estimator although it does not consistently estimate the distribution of the MLE. A Monte Carlo experiment shows that bootstrap-based confidence intervals that rely on the half-panel jackknife estimator greatly help to reduce the distortions associated to the normal approximation in finite samples. In addition, I apply this bootstrap method to a canonical model of female-labor participation to construct valid confidence intervals. In the last chapter (co-authored with Wenjie Wang), we study the asymptotic validity of bootstrap procedures for instrumental variable (IV) models with many weak instruments. We show analytically that a standard residual-based bootstrap and the restricted efficient (RE) bootstrap of Davidson and MacKinnon (2008, 2010, 2014) cannot consistently estimate the limiting distribution of the LIML estimator. The foremost reason is that they fail to adequately mimic the identification strength in the sample. Therefore, we propose a modified bootstrap procedure which consistently estimates this limiting distribution. Our simulations show that the modified bootstrap procedure greatly reduces the distortions associated to asymptotic Wald ($t$) tests in finite samples, especially when the degree of endogeneity is high.