Financial constraints and limited access to markets are the most important obstacles for economic development in developing economies that are largely dependent on agriculture. Lack of access to credit and output markets, in particular, is frequently identified as a key barrier to transformation of subsistence agriculture. The fundamental problems are related to information asymmetry, lack of collateral and limited economies of scale. Innovative institutional mechanisms, in the form of microfinance and producer organizations, offer ways to address information asymmetry and increase scale economies. This dissertation examines the outreach, financial performance and impact of microfinance institutions as well as the key drivers of agricultural cooperatives and its impact on smallholder farmers efficiency in Ethiopia. To meet these objectives four data sets from Ethiopia, one institutional and three household survey data, are used. The analysis of microfinance generally focuses on examining the trade-offs between outreach to the poor and financial sustainability by ownership forms (shareholder owned vs. memberowned). The outreach-financial sustainability trade-offs analyzed first using the institutional survey data. The empirical results show that serving the poor and financial sustainability are challenging objectives to achieve together. There is also evidence that suggests the presence of mission drift. Second, the role of ownership form on outreach, financial performance and cost-efficiency is analyzed within the framework of efficient ownership assignment theories using regression and stochastic cost frontier approaches. This analysis aims at testing whether the commonly held proposition of greater efficiency of shareholder firms in microfinance by policy advocates is empirically supported. The results reject the superiority of shareholder-owned microfinance over financial cooperatives. In fact, the evidence here supports the advantages of member-owned microfinance on cost efficiency and balancing the double bottom-lines of microfinance. Third, the impact of access to microfinance credit on farmers investment on agricultural inputs is assessed using propensity score matching (PSM) and control-function-regression methods that address potential participation selection biases. Results from both approaches show that access to credit increased the use of productivity enhancing inputs among borrower farmers. Indeed, farmers that borrow from financial cooperatives tend to invest more on modern inputs. The analysis of agricultural cooperatives tested theoretical propositions from organizational and the new institutional economics theories on the drivers of agricultural cooperatives incidence and farmers membership and patronage decisions. Discrete choice models are employed for the empirical analysis. The results indicate that the incidence of agricultural cooperatives in Ethiopia is more related to the countervailing market power argument than to the transaction cost reduction hypothesis. Despite open membership polices, the analysis on farmers participation indicates that membership and patronage decisions are related and significantly influenced by location, asset and relational specificities. Finally, the impact analysis estimated technical efficiency gains of membership in agricultural cooperatives and employed matching and stochastic production frontier techniques. The evidence suggests that membership in agricultural cooperatives significantly enhances efficiency gains among smallholder farmers.

The overall goal of this work is to study the effect of a crisis on the distribution and employment and the space of manoeuvre of the government for supporting and reverting the negative shock produced by such a crisis. Every chapter of this work and the related models are supported by both a theoretical background analysis and by numerical dynamic simulations. Stylized facts show that the income and wealth inequality in all the OECD countries has been constantly increasing after the 1960s. Piketty has been one of the most important authors that highlighted the rising inequality issue, mainly in the OECD countries. For example, Piketty (2014) shows that the income share held by the top percentile in countries such as US, Canada and UK increased from 8%-10% in the 1960s up to 14%-18% in the current decade. Similar figures are now provided by the World Inequality Lab that has updated data for almost every country up to 2016. At the same time, the wage share for the majority of the OECD countries substantially decreased. For example, countries such as Italy and Spain experienced a decrease in wage share from about 73% in the 1970s to about 63% in the current decade (Hein, 2014). Taking into account such a stylized fact, we will consider a model with two social classes, workers and capitalists. These social classes differ in terms of their initial endowment, their consumption behaviour, the different loans repayment conditions required to them by the banks and in terms of the ways in which they can use their financeable wealth. This is a very important departure hypothesis from the mainstream point of view models that generally consider a population made up of “a representative agent” of the whole society. Considering the inequality levels that the OECD countries are experiencing, we took the Post-Keynesians school of thought as a very good reference point since it always focused its attention on the relation between the level of employment, the aggregate demand and the distribution between social classes. In line with the post-Keynesians tradition, we believe that a theory cannot be correct unless it starts from realist or realistic hypotheses, although it is recognized that assumptions are always abstractions and simplifications (Lavoie, 2014). Therefore, we developed a step by step model with the analysis of an economy based on some well-known stylized facts. Beyond the social classes distinction, we take into consideration the temporal lag between production and sales of products by firms and the one between income received by the social classes and their expenditure. Those two temporal lags are the very key aspects we focus our attention on in the model presented in Chapter II named “Keynes, Kalecki and Metzler in a Dynamic Distribution Model”. In that chapter, we merge the hints of Keynes and Kalecki about the distribution of social classes and the intervention of the government in supporting the aggregate demand together with Metzler’s hint about the mismatching process between aggregate demand and aggregate production. Metzler’s mismatching process would finally generate inventories of consumption goods. More specifically, it is argued that even if Post-Keynesians models focused their attention on output growth, employment and income distribution relating those issues with a stronger intervention of the state, they all (even the canonical Kaleckian model) overlooked the adjustment - or non-adjustment - dynamics from the ultra-short run to the short run period upon which the short run and long run models are then constructed. In fact, even if the Kaleckian models completely reject the standard neoclassical production function (rejecting diminishing returns and rejecting the substitution between capital and labour) they also very strongly rely on a final equilibrium between aggregate demand and aggregate production. The canonical Kaleckian short run models are constructed upon the consideration of the effective labour demand curve defined as “the locus of combinations between real wages and levels of employment which ensure that all produced goods are sold at the price set by firms” (Lavoie, 2014). As argued by Lavoie (2014), this construction assures that an increase of real wage leads to an increase in the employment level. That has been and still is definitely one of the cornerstones for the Post-Keynesian authors. We argue that the equilibrium assumption between the aggregate demand and the aggregate production plays a key role in obtaining the standard Kaleckian conclusions regarding the relation between effective demand, employment levels and the distribution of surplus product between the social classes. The main question arising from the previous enquiring exercise about adjustment dynamics in the Kaleckian framework is that, because of the overlooking on that adjustment process between aggregate production and aggregate demand, also its conclusions might be consequentially affected. More precisely the main Post-Keynesian Kaleckian conclusions to assess are the following: would it still be true that higher real wages lead to a higher level of employment? Would it still be true that a decrease in the propensity to save will lead to an increase in output and employment? Would it still be true that in order to keep employment from falling, whenever there is an increase in productivity there must be some increase in real wages? And finally, most importantly in terms of policies, would it still be true that in order to keep employment from falling, even when the economy faces a pari passu increase of real wage and productivity level, it would be necessary an increase in real autonomous expenditure such as a strong government one? In this way, our model analyses under which conditions the standard Kaleckian conclusions are still valid considering a disequilibrium situation. Two scenarios are simulated: one with fixed expectations as in Metzler (1941) and another new one based on adaptive expectations and asymmetric behaviour of the wages-unemployment relation. The model questions the effective demand labour curve and suggests that an increase in real autonomous expenditures, mainly by the Government, might be even more essential than what is generally considered in the Kaleckian literature, to avoid increasing unemployment in an increasing wage world. The model presented in Chapter III named “The stabilising role of the Government in a Dynamic Distribution Growth Model” builds upon the model presented in Chapter II and considers once again the effect of a crises on the relation between aggregated demand, employment and distribution between social classes adding important characteristics of realism that were absent in the previous chapter. Here, we consider the gestation period of the investments and the presence of the government investigating its margin of manoeuvre in such an economy. The first aspect takes inspiration by Kalecki (1971) himself who considers the three different Investment stages: investment order or Demand (I^D), investment Production (I^P) and investment delivery or Completion (I^C). In line with a post-Kaleckian perspective, we consider the expected profitability and the capacity utilisation as the two main variables as driving forces for the investment decisions. The second new aspect of this model compared to the one presented in Chapter II is the explicit presence of the government. In fact, even if chapter II suggested the Government as the emblematic autonomous figure able to foster expenditure in times of recession, its actual role in the economy was not analysed. Many post-Keynesian scholars have underlined how recent decades have been characterised by a strong downgrading of the fiscal policy role as a stabilisation instrument of macroeconomic policy (Arestis and Sawyer, 2003). In this way, this chapter analyses exactly the space of manoeuvre of the government and the role of the fiscal policies into a “functional finance” framework where the government "can and should be called upon as a key part of the remedy" (Fazzari, 1994) to ensure a high level of economic activity whenever the private sector is unable to do so by itself. In the light of such a functional finance framework, the government actions should be inspired to achieve a more stable and sustainable growth path. More specifically, we here investigate the possibilities that the Government has to boost and support the economic activity with its two main tools, public investments spending and a taxation system in two scenarios. The first scenario simulates an exogenous fall of private investments while the second one relates to an exogenous increase in labour productivity and real wages. In particular, here we test the canonical Kaleckian model conclusion according to which even when the economy faces a pari passu increase of real wages and productivity level it would be necessary an increase in real autonomous expenditures - such as the one implemented by the government - in order to keep employment from falling. At the same time, the aim of this chapter is also to explore the role of the Government in stabilising the economy exactly thanks to the previous tools. In fact, Chapter II underlined the possibility of an arising unstable path from a mismatching dynamic between aggregate demand and aggregate production. It was argued that such an unstable path might develop because of “wrong” oversensitive expectations of firms regarding the production of consumption goods. Therefore, chapter III focuses exactly on the space of manoeuvre of the government in stabilizing an unstable economic scenario caused by a crisis. The model built in Chapter IV named “The distributive monetary analysis of a sustainable ecological economy” is the natural evolution of the models developed in Chapters II and III. In such a model all the previous stylized facts are contained, namely the temporal lag between production and sales of products by firms, the temporal lag between income received by the social classes and their expenditure, the gestation period of the investments and, finally, the intervention of the government. The most important difference with respect to the models presented in the previous chapters is its overall monetary and ecological framework. In fact, for simplification purposes the previous models were assuming that, in line with a horizontalist approach, commercial banks were providing funds on demand to firms for financing their investments. However, the explicit relations among all the sectors of our economy were not fully exposed. In this chapter Graziani’s endogenous money theory is used and we are developing a Post-Keynesian Stock Flow Consistent (SFC) model to track all the economic relations, both the real and monetary ones. At the same time, the use of a SFC model ensures that “there are no black holes - every flow comes from somewhere and goes somewhere” (Godley W. , 1996) through a rigorous accounting framework, which guarantees a correct and comprehensive integration of all the flows and the stocks of an economy. Such as Kalecki, Graziani and the circuitists economists introduce a preliminary distinction between producers and wage earners. The first step of the monetary circuit is always characterized by firms’ decision to activate production and, in order to do so, they take up loans by commercial banks. In this sense, commercial banks are able to create deposits ex nihilo, granting them loans and, at the same time, creating deposits. In this way, the starting logical cause of the expansion of money is exactly the firms’ willingness of contracting a liability to activate production. In the second step, firms use those loans to pay workers and in this way to obtain the amount of consumption goods desired through the production process. When such funds are transferred by firms to households they instantaneously become income paid for the work provided to firms by workers. Finally, the last step of the Monetary Circuit is characterized by the households’ spending decision to use the money balances previously obtained as income. In this step, while households use their funds to buy consumption goods, firms obtain back those money balances they initially paid to households for their work. In this way, the previous Monetary Circuit analysis is not in contrast with the one made by Kalecki upon the way workers obtain their wages and use all of them to buy consumption goods while capitalists are able to spend just a proportion of their income. Finally, together with its social and monetary framework, our economy is also characterized by an environmental one since we here study the impacts that the economic consumption has in terms of ecological erosion of natural resources. In this way, the model of chapter IV questions the expenditure margins of the Government – in particular after a crisis - and uses the suggestions of the monetary circuit theory to analyse the space for fiscal policies to reduce unemployment boosting the economic activity, to obtain a more equitable distribution between social classes in a sustainable ecological way. Our understanding is that despite many contributions focused on the topics of recovery, distribution and ecological sustainability, few of them tried to tackle them all in a comprehensive way considering the rediscovery of the endogenous money phenomena as one of the most important breakthroughs in the last decades. Here we argue that exactly the endogenous money feature is the essential fil rouge to better understand and connect the three previous important aspects. It is so when we analyse the sectors connections and the policies ones devoted to recovery, and also if we consider how the different incomes and wealth are captured and distributed by the different social classes and finally when we point out the ways of financing long term ecological path to preserve a sustainable environment. Indeed, our overall work in Chapter II, Chapter III and Chapter IV is a step by step construction of an organic and consistent model. It starts with a more theoretical and simplified approach through Chapter II which investigates the (in)stability conditions of the Kaleckian approach while suggesting the presence of an autonomous figure such as the government one. Chapter III adds more real base features through endogenous investments and government presence while Chapter IV finally concludes considering all the real and monetary links of the sectors into a social and ecological framework.

“What drives firms' performance?” remains one of the most intriguing topics in the economic literature. The myriad determinants identified so far include both internal and external factors. While the internal factors tend to relate to either firms' characteristics or the business decision-making process determined by managers/owners, the external factors tend to relate to the business environment and any possible stimulus for the conditions in which firms are located and operate. This thesis follows the strand of literature that continues the quest for relevant firm-performance-driving factors, taking an applied approach. The thesis, consisting in three essays, provides empirical evidence of how trade liberalization, together with local business environment features, affects firms' performance in the context of a transitional country, Vietnam. The main focus centres particularly on firms' productivity (the first and second essays). As firms use labour as an indispensable input of production, interest, in the third essay, shifts to firm-level demand for skills so as to understand also potential consequences of trade reform policies for the labour market. Highlights of the three essays are as follows: The first essay (Chapter 2) investigates the interplay of trade liberalization and local business environment features in affecting firms’ productivity. Particular interest lies in how frictions in the local business environment where firms operate alter the productivity gains from trade. Making use of a large panel dataset of Vietnamese manufacturing firms from 2006 to 2012, the analysis provides robust evidence of a positive effect of trade liberalization on firms’ total factor productivity. However, distortions in the local business environment such as unenforceable property rights, an ineffective land-titling system, bureaucratic hurdles and labour market frictions play a crucial role in the transmission of trade liberalization shocks. It is recommended, from the results obtained, that complementary business environment reforms addressing local market constraints need to be implemented alongside trade liberalization. The second essay (Chapter 3) investigates first the impact on firms’ productivity of trade liberalization and second how corruption interacts with trade policies. Particular interest centres on how firms’ productivity improvement induced by trade openness is reduced by corruption. Using firm-level data on Vietnamese manufacturing enterprises from 2000 to 2012, the period of which spans the country's accession to the World Trade Organization (WTO), the analysis provides robust causal evidence that trade openness increases firms’ productivity, the higher so if firms export. However, the productivity gains expected by firms from trade liberalization are reduced by corruption, which has important implications for aggregate productivity. The findings suggest that complementary policies addressing either export constraints or corruption need to be implemented in company with trade reform policies. The third essay (Chapter 4) investigates the firm-level employment consequences of trade liberalization. Particular interest is given to how small and medium-sized enterprises (SMEs) adjust their labour demand in response to different types of trade shocks. Using a unique dataset of private manufacturing SMEs over the 2003--2014 period that captures the 2007–2008 drastic reduction in both output and input tariffs, that analysis shows that the effects on firms' labour demand of output and input tariff liberalization are non-identical. While output tariff liberalization has a labour-friendly effect on unskilled workers, a reduction in input tariffs has an “absolute” skill-biased effect, shifting firms’ demand from unskilled workers to skilled labour. The findings suggest important consequences of trade openness for the labour market evolution. A cautious view on the employment consequences of trade openness is strongly suggested, coupled with the implementation of coherent education and training policies to guide the labour supply under trade liberalization. The contributions of this thesis to the literature are as follows. First, it helps to understand how firms adjust to trade liberalization policies in terms of both behaviour and performance in the context of a developing country in transition. Second, it shows that the effectiveness of trade liberalization policies significantly varies with the business environment conditions, explaining why gains from trade are unevenly distributed across locations. Third, in a movement towards free trade, benefits are shown to go hand in hand with drawbacks. In terms of policy implications, a thorough multi-dimensional view on the potential consequences of trade openness is highly recommended. Besides, the effective implementation of appropriate complementary policies in the wake of trade liberalization is of utmost importance in order to first achieve an overall optimal economic outcome and second make trade openness beneficial for everyone.