Essays on finance and macroeconomy

I study the asset-pricing implications in an cnviromncnt with feedback traders and rational arbitrageurs. Feedback traders are defined as possible naive investors who buy after a raise in prices and sell after a drop in prices. I consider two types of feedback strategies: (1) short-term (SF), motivated by institutional rulcs as top-losscs and margin calls and (2) long-tcrm (LF), motivated by representativeness bias from non-sophisticated investors. Their presence in the market follows a stochastic regime swift process. Short lived assumption for the arbitrageurs prevents the correction of the misspricing generated by feedback strategies. The estimated modcl using US data suggests that the regime switching is able to capture the time varying autocorrclation of returns. The segregation of feedback types helps to identify the long term component that otherwise would not show up due to the large movements implied by the SF typc. The paper also has normativo implications for practioners since it providos a methodology to identify mispricings driven by feedback traders.

Demand aggregation and credit risk effects in pooled procurement: evidence from the Brazilian public purchases of pharmaceuticals and medical supplies

Pooled procurement has an important role in reducing acquisition prices of goods. A pool of buyers, which aggregates demand for its members, increases bargaining power and allows suppliers to achieve economies of scale and scope in the production. Such aggregation demand e ect lowers prices paid for buyers. However, when a buyer with a good reputation for paying suppliers in a timely manner is joined in the pool by a buyer with bad reputation may have its price paid increased due to the credit risk e ect on prices. This will happen because prices paid in a pooled procurement should refect the (higher) average buyers' credit risk. Using a data set on Brazilian public purchases of pharmaceuticals and medical supplies, we nd evidence supporting both e ects. We show that the prices paid by public bodies in Brazil are lower when they buy through pooled procurement than individually. On the other hand, federal agencies (i.e. good buyers) pay higher prices for products when they are joined by state agencies (i.e. bad buyers) in a pool. Such evidence suggests that pooled procurement should be carefully designed to avoid that prices paid increase for its members.

Forecasting multivariate time series under present-value-model short- and long-run co-movement restrictions

It is well known that cointegration between the level of two variables (e.g. prices and dividends) is a necessary condition to assess the empirical validity of a present-value model (PVM) linking them. The work on cointegration,namelyon long-run co-movements, has been so prevalent that it is often over-looked that another necessary condition for the PVM to hold is that the forecast error entailed by the model is orthogonal to the past. This amounts to investigate whether short-run co-movememts steming from common cyclical feature restrictions are also present in such a system. In this paper we test for the presence of such co-movement on long- and short-term interest rates and on price and dividend for the U.S. economy. We focuss on the potential improvement in forecasting accuracies when imposing those two types of restrictions coming from economic theory.

Speak now or forever hold your peace: an empirical investigation of whistleblowing in Brazilian organizations

À medida que ajudam as organizações a prevenir condutas irregulares e a manter um ambiente de trabalho mais éticos, as denúncias (whistleblowing) são frequentemente apontadas pela literatura como um benefício à sociedade como um todo. Entretanto, pouco se sabe a respeito de muitos aspectos associados à decisão de denunciar uma irregularidade. No Brasil, em especial, onde o assunto permanece sendo negligenciado por pesquisadores, elementos culturais específicos podem obstruir o caminho da denuncia, além de impor restrições à generalização dos resultados de pesquisas anteriores, quase sempre voltadas para a realidade anglo-saxônica. Com base nesses pressupostos, este estudo busca identificar os antecedentes do whistleblowing interno nas organizações brasileiras. De modo geral, os resultados da pesquisa empírica realizada com uma ampla amostra de profissionais oriundos de empresas púbicas e privadas dão suporte ao modelo proposto e reforçam a noção de que o ato de denunciar é o resultado complexo da interação entre fatores da organização, do indivíduo da situação observada. Em particular, os resultados sugerem que os indivíduos são mais propensos a realizar a denúncia quando a irregularidade observada é percebida como grave e quando a própria denuncia é vista como um curso de ação ético. Por outro lado, os indivíduos podem optar pelo silêncio se o alvo da denúncia for alguém com alto status na organização, se eles não sentirem apoio da organização ou se temerem retaliações. A influência negativa do medo de retaliação sobre a intenção de denunciar, contudo, pode ser atenuada se o indivíduo está convencido de que o correto a fazer é denunciar. Por fim, os resultados também sugerem que profissionais em posição gerencial são mais propensos a denunciar irregularidades do que os demais membros de uma organização. As implicações desses resultados, bem como as limitações e contribuições do estudo para a teoria e para a prática são discutidas em detalhe, juntamente com sugestões para pesquisas futuras.

Health and development with imperfect capital markets

This paper investigates the interaction between investment in education and in life-expanding investments, in a simple two-period model in which individuaIs are liquidity constrained in the first period. We show that under low leveIs of health and capital, investments in human capital and in health are complement: since the probability of survival is small, there is littIe incentive to invest in human capital; therefore the return on health investment is also low. This reinforcing effect does not hold for higher leveIs of health or capital, and the two investments become substitute. This property has many consequences. First, subsidizing health care may have dramatically different effects on private investment in human capital, depending on the initial leveI of health and capital. Second, the assumption that mortality is endogenous induces an increase in inequality of income: since health investment is a normal good, the return on education is also lower for poor individuaIs. Third,in a non-overlapping generation madel with non-altruistic agents, the hea1th leveI of the population has strong consequences on growth. For a very low leveI of hea1th, mortality is too high for the investment on education to be profitable. For a higher, but still low, levei of hea1th the economy grows on1y if the initial stock of capital is high enough; bad health and low capital create a poverty trapo Fourth, we compare redistributive income policies versus public hea1th measures. Redistributing income reduces both static and dynamic inequality, but slows growth. In contrast, a paternalistic health policy that forces the poor to invest in hea1th reduces dynamic inequality and may foster growth.

Uncertainty of aid inflows and the growth relationship

This paper contributes to the literature on aid and economic growth. We posit that it is not the levei of aid flows per se but the stability of such flows that determines the impact of aid on economic growth. Three measures of aid instability are employed. One is a simple deviation from trend, and measures overall instability. The other measures are based on auto-regressive estimates to capture deviations from an expected trend. These measures are intended to proxy for uncertainty in aid receipts. We posit that such uncertainty will influence the relationship between aid and investment and how recipient governments respond to aid, and will therefore affect how aid impacts on growth. We estimate a standard cross-country growth regression including the leveI of aid, and find aid to be insignificant (in line with other results in the literature). We then introduce measures of instability. Aid remains insignificant when we account for overall instability. However, when we account for uncertainty (which is negative and significant), we find that aid has a significant positive effect on growth. We conduct stability tests that show that the significance of aid is largely due to its effect on the volume of investment. The finding that uncertainty of aid receipts reduces the effectiveness of aid is robust. When we control for this, aid appears to have a significant positive influence on growth. When the regression is estimated for the sub-sample of African countries these findings hold, although the effectiveness of aid appears weaker than for the full sample.

The effects of uncertainty about the timing of deals on consumer behavior

I examine the effects of uncertainty about the timing of de aIs (i.e. temporary price cuts or sales) on consumer behavior in a dynamic inventory model of consumer choice. I derive implications for purchase behavior and test them empirically, using two years of scanner data for soft drinks. I fmd that loyal consumers' decisions, both about the allocation of their purchases over time and the quantity to be purchased in a particular deal, are affected by the uncertainty about the timing of the deal for the product. Loyal consumers buy a higher fraction of their overall purchases during de ais as the uncertainty decreases. This effect increases with an increase in the product' s share of a given consumer' s purchase in the same category or if the consumer stockpiles (i.e., is a shopper). During a particular deal, loyal shoppers increase the quantity they purchase the more time that has passed since the previous de aI, and the higher the uncertainty about the deals' timing. For the non-Ioyal consumers these effects are not significant. These results hold for products that are frequently purchased, like soft-drinks and yogurt, but do not hold for less frequentIy purchased products, such as laundry detergents. The fmdings suggest that manufacturers and retailers should incorporate the effects of deals' timing on consumers' purchase' decisions when deriving optimal pricing strategies.

Buy-or-sell auctions

This paper develops a game theoretic model of a "Buy-or-Sell" auction. Participants have to submit both a bid and an offer price for up to one of the many units of the good being auctioned. The bid-ask spread is set in advance by the auctioneer. Such an auction was used by the Central Bank of Brazil to intervene in the foreign exchange market during the exchange rate crawling-peg regime (1995-1999). I investigate whether such mechanism is more effective than standard intervention auctions to prevent speculative attacks in the context of managed exchange rate regimes.

Cabinet composition and assessment of a multiparty presidential system

This doctoral dissertation provides a detailed analysis of the Brazilian cabinet according to the concepts of a multiparty presidential system. Appointing politicians as ministers is one of the most important coalition-building tools and has been widely used by minority presidents. This dissertation will therefore analyze the high-level Brazilian national bureaucracy between 1995 and 2014. It argues that the ministries – or departments – are not equal, and that allied parties therefore take into account the different characteristics of a ministry when demanding positions as a patronage strategy or for use as other kinds of political assets. After reviewing the literature on the theme, followed by a comparative analysis of the Brazilian, Chilean, Mexican, and Guatemalan cabinets, all the Brazilian ministries will be weighed and ranked on a scale that is able to measure their political importance and attractiveness. This rank takes into account variables such as the budgetary power, the ability to spend money according the ministers’ will, the ability to hire new employees, the ministries’ influence over other governmental agents such as companies, agencies, and so on, the ministers’ tenure in office. Finally, a proxy is provided that seeks to identify the normative power a department may hold. All of these characteristics will then be taken into account in considering the representatives’ opinion, thus helping to ascertain whether the cabinet appointment has been coalescent among the several parties that belong to the president’s coalition.

The spenders-hoarders theory of capital accumulations, wealth distribution and fiscal policy

This paper proposes a simple OLG model which is consistent with the essential facts about consumer behavior, capital accumulation and wealth distribution, and yields some new and surprising conclusions about fiscal policy. By considering a society in which individuais are distinguished according to two characteristics, altruism and wealth preference, we show that those who in the long run hold the bulk of private capital are not so rnuch motivated by dynastic altruism as by preference for wealth. Two types of social segmentation can result with different wcalth distribution. To a large extcnt our results seem to fit reality better than those obtained with standard optimal growth models in which dynastic altruism ( or r ate o f impatience) is the only source of heterogeneity: overaccumulation can appear, public debt and unfunded pensions are not neutra!, estate taxation can improve the welfare of the top wealthy.