ABSTRACT. This paper provides a general method to directly translate a classical economic framework with a large number of agents into a field-formalism model. This type of formalism allows the analytical treatment of economic models with an arbitrary number of agents, while preserving the system's interactions and microeconomic features of the individual level. We apply this methodology to model the interactions between financial markets and the real economy, described in a classical framework of a large number of heterogeneous agents, investors and firms. Firms are spread among sectors but may shift between sectors to improve their returns. They compete by producing differentiated goods and reward their investors by paying dividends and through their stocks' valuation. Investors invest in firms and move along sectors based on firms' expected long-run returns. The field-formalism model derived from this framework allows for collective states to emerge. We show that the number of firms in each sector depends on the aggregate financial capital invested in the sector and its firms' expected long-term returns. Capital accumulation in each sector depends both on short-term returns and expected long-term returns relative to neighbouring sectors. For each sector, three patterns of accumulation emerge. In the first pattern, the dividend component of short-term returns is determinant for sectors with small number of firms and low capital. In the second pattern, both short and long-term returns in the sector drive intermediate-to-high capital. In the third pattern, higher expectations of long-term returns drive massive inputs of capital. Instability in capital accumulation may arise among and within sectors. We therefore widen our approach and study the dynamics of the collective configurations, in particular interactions between average capital and expected long-term returns, and show that overall stability crucially depends on the expectations' formation process. Expectations that are highly reactive to capital variations stabilize high capital configurations, and drive low-to-moderate capital sectors towards zero or a higher level of capital, depending on their initial capital. Inversely, low-to moderate capital configurations are stabilized by expectations moderately reactive to capital variations, and drive high capital sectors towards more moderate level of capital equilibria. Eventually, the combination of expectations both highly sensitive to exogenous conditions and highly reactive to variations in capital imply that large fluctuations of capital in the system, at the possible expense of the real economy.

ABSTRACT. The aim of this paper is to propose a new model of bubbles and crashes in order to elucidate a mechanism of bubbles and subsequent crashes. We consider an asset market where risky assets are traded in two classes, the risky asset and the risk-free asset. Investors can be divided into two groups, each with a different rationale for making decisions. One is the fundamentalists, who maximise their expected utility of their wealth in the next period following their rational assessment of the fundamental values of risky assets. The other is speculators, who maximise their random utility of a binary choice: buy the bubble asset and hold the risk-free asset. The speculator's behaviour is modelled in the framework of the Ising spin model of statistical mechanics, which can be seen as a model of Keynse's beauty contest metaphor. We show that (i) if the speculators' conformity effect (the extent to which each noise trader is influenced by the decisions of other speculators) is weak, then the market price converges to the fundamental price, so that the efficient market hypothesis holds, but that (ii) if the speculators' conformity effect is strong, then speculators' herding behaviour leads to a bubble, and their positive-feedback trading prolongs the bubble, but a bubble necessarily ends in a crash. Furthermore, we describe that cycles of bubbles and crashes are repeated.

ABSTRACT. In finance, information moves prices through the interaction of market participants. Information and interaction are two sides of the same coin. Market efficiency provides a convenient shortcut that made it possible to build the edifice of mathematical finance without taking interaction into account. Many agent based models have addressed the issue of how information is incorporated into prices, providing key insights on market microstructure. Yet finance still has no solid information theoretical foundation.

I will revisit the Glosten-Milgrom model to dig deeper in the relation between information and interaction. First, I will show that the main empirical findings on the impact of a sequence of transactions on prices of financial assets can be recovered within a surprisingly simple version of the Glosten-Milgrom model with fully Bayesian market expectations. These empirical findings, which are among the most well established results in market macrostructure, show that the price impact, i.e. the expected change in price due to a sequence of trades of total volume $Q$, is proportional to the square-root of $Q$. Furthermore, for very small values of $Q$, the square-root behaviour crosses over to a linear regime. Finally, the impact decays back to equilibrium, after the sequence of trades is over. All of these three stylized facts are recovered within a very simple framework. The simplicity of this derivation lends further support to the robustness and universality of the empirical findings, and it sheds light on its origin. In particular, it suggests that the square-root impact law originates from the over-estimation of order flows (more details at L. Saddier and M. Marsili, arXiv:2303.08867).

Second, I will show that, within the Glosten-Milgrom model, the maximal expected gain that informed traders can extract from the market is bound by the same relation that governs the maximal work that can be extracted from a cyclic thermodynamic transformation, when some information on the microscopic state of the physical system is known. This is done deriving an analogy between the Glosten-Milgrom model and a thermodynamic (Szil\'ard) information engine, whereby the optimal work extraction protocol in the latter coincides with the pricing strategy of the market maker in the former. Furthermore, we show how the analogue of the physical temperature can be defined in the Glosten-Milgrom model, from the analysis of the distribution of market orders. Finally we show that the expected gain of informed traders is bounded above by the product of this market temperature with the amount of information that informed traders have, in exact analogy with the corresponding formula for the maximal expected amount of work that can be extracted from a cycle of the information engine. Overall, these results suggest a strong analogy between market efficiency and the second law of thermodynamics, and hat recent ideas from information thermodynamics may paves the way for an information theoretic approach to finance (more details at L. Touzo, M. Marsili and D. Zagier, {\em Information thermodynamics of financial markets: the Glosten-Milgrom model}. JSTAT 2021.3 (2021): 033407).

ABSTRACT. In this paper, we investigate the implications of temporal aggregation, i.e. the discrepancy between the Data Generation Process (DGP) and the Data Collection Process (DCP), for the design of monetary policy in a New Keynesian macroeconomic framework with boundedly rational agents. We extend the model by Airaudo, Nisticó and Zanna (2015) who investigate if monetary policy should explicitly respond to stock prices due to the presence of a structural linkage between the stock market and the real activity. We stress this rationale in a similar model with agents with heterogeneous boundedly rational expectations by showing that responding to the stock price is further justified when real data is only available at a delay due to temporal aggregation. Under these conditions, moderately reacting to high frequency stock price movements can stabilize both the financial and the real sectors.

ABSTRACT. While the rational choice approach has been the theoretical modeling paradigm in political science over the past decades, the importance of behavioral factors such as heuristics and biases has been being increasingly acknowledged, see e.g. Bendor (2010) and Bendor et al. (2011). For instance, Di Guilmi and Galanis (2021) develop a behavioural dynamic model with heterogeneous individuals with endogenously evolving preferences (in the Brock-Hommes tradition) and two policy-oriented parties. The parties have different core values in terms of income redistribution but the extent to which their policies are actually implemented depends on the relative support that they enjoy. Their policies change the income distribution and generate a feedback effect on electoral preferences. As the dynamics of income distribution and the macroeconomy in general Di Guilmi and Galanis (2021) are quite parsimoniously specified, we aim in the proposed paper to integrate it with the Keynes-Metzler-Goodwin (KMG) framework (see e.g. Chiarella et al., 2005, Chiarella et al., 2021) in order to model these dynamics, and in general the evolution of the macroeconomy, in a more structured manner, and to investigate how party politics may interact with macroeconomic dynamics in such a theoretical framework.

ABSTRACT. The expenditure cascades hypothesis, that income gains at the top of the income distribution lead to greater expenditures throughout the distribution, is typically built on upward-looking social comparisons, the idea that individuals only compare themselves to those in a higher income quantile. I argue that the assumption of only upward-looking comparisons is spurious and unnecessary. I propose an agent-based model of consumption decisions at the micro-level with social comparisons mediating in an explicit perception network, which employs a novel method of integration of the consumption reference into the utility function. Using the model, I demonstrate that expenditure cascades can occur without the upward-looking assumption, but that dropping the assumption has dramatic implications for the nexus of inequality and aggregate consumption, and whether Keynesian or cascade effects dominate. These observations offer a new theoretical perspective on the ambiguous empirical evidence in this area. I also make similar observations about the effects of income segregation in the network on aggregate consumption. The shape of expenditure cascades has important implications for macro- and ecological economics, as it helps us to understand the transmission process from policies affecting inequality to changes in total consumption.

ABSTRACT. Economic History(EH) is a peculiar discipline. Emerging from the encounter of one of the oldest disciplines, History, with one of the newest, Economics, Economic History has never managed to mend this dichotomy fully. To this day, its practitioners divide themselves by their respective backgrounds: historians perform qualitative work, and economists specialize in quantitative methodologies. They collaborate, of course, but neither camp can offer a syncretic approach capable of bringing together the two souls of the discipline. Agent-Based Models (ABMs) are a recent class of models. Despite this, ABMs see use in a variety of natural and social sciences. For EH, ABMs offer a flexible modelization technique compatible with historical sources, as demonstrated by its ample use in Archeology, and AMBs can also help understand and reproduce economic phenomena, as shown by evolutionary and Schumpeterian economics. The adoption of ABMs by economic historians could provide them with a methodology able to build credible virtual histories based on historical evidence, effectively allowing the use of counterfactuals for a better understanding of case studies and theories. This paper is dedicated to building the case for this approach. To do so, we provide a brief introduction to counterfactuals in social sciences, followed by the failures of standard modelization in providing satisfying counterfactual to Economic History. It then continues with an introduction to Agent-Based modelization, followed by an exploration of the History-Friendly models, their motivation, and what can be improved to provide the discipline with credible counterfactuals.

ABSTRACT. Incentive contracts are widely employed to affect subordinates' behavior in hierarchical organizations and induce effort to achieve the organization's overall objectives. In this vein, principal-agent theory regards incentive contracts as a primary means to deal with problems of adverse selection and moral hazard (e.g., Eisenhardt, 1989; Lambert, 2001). However, incentive contracting encounters various difficulties, such as measuring the performance in all task elements assigned to a subordinate or not foreseeing all future states of the environment (e.g., Hart, 1988, 2017; Tirole, 1999). These difficulties may result in incomplete incentive contracts. Contractual incompleteness in terms of incentive contracts not governing all task elements for which a subordinate is responsible is a ubiquitous and relevant phenomenon: Many tasks comprise elements like innovativeness, creativity, courtesy, or empathy for others and the environment that are hard to measure; at the same time, these elements may remarkably affect an organization's success (e.g., Holmström & Milgrom, 2009; Kreps, 1990; Sanga, 2018). A broad stream of prior research studies incomplete incentive contracts in organizations. In the tradition of principal-agent theory, incomplete incentive contracting typically was examined as a multi-task problem. In this vein, building on assumptions of “far-reaching” actors’ rationality, the conditions were studied for when additional performance metrics or particular job designs and job allocation are helpful (e.g., Dewatripont, Jewitt, & Tirole, 2000; Feltham & Xie, 1994). The role of trust in incomplete contracts is at the center of research of a different body of research that predominantly focuses on the subordinate’s trust in the superior (e.g., Christ, Sedatole, & Towry, 2012; Fisher, Maines, Peffer, & Sprinkle, 2005): The background is that with an incomplete contract, the superior has some discretion over the subordinate’s rewards and, in consequence, the subordinate’s trust in the superior is particularly relevant for the subordinate’s willingness to incur effort related to the not-contracted task elements. However, it is well noticed that trust is “two-sided,” i.e., that also the superior’s trust in the subordinate is to be considered (with further references, Schoorman, Mayer, & Davis, 2007) This paper translates these insights into an agent-based study of incomplete incentive contracts in hierarchical organizations. In particular, the paper seeks to provide a deepened understanding of the emergence of reciprocal trust and the performance effects thereof in settings of incomplete incentive contracts. In this sense, the paper picks up indications of the relevance of trust-building highlighted in the economics of reciprocity (e.g., Fehr & Gächter, 2000; Sethi & Somanathan, 2003). For this, an agent-based simulation model based on NK fitness landscapes (e.g., Kauffman & Levin, 1987)is employed, extending the model introduced in Wall (2022). The extensions refer to the inclusion of trust and the reciprocal emergence of trust among parties. For the latter, the study builds on the integrated model of Brower, Schoorman, and Tan (2000) as a theoretical basis for trust-building in organizations. In the simulation experiments, the complexity of the task environment and the propensity for trust-building on the superior's side are varied. The results suggest that trust on the superior’s side neither universally increases with the superior’s propensity to reciprocate the perceived trust from the subordinates nor does organizational performance or the relative not-contracted performance increase generally with higher trust levels. Among the reasons is that, with the superior’s imperfect performance measurements, the superior’s trust in the subordinate might be “unjustified” from a subordinate’s perspective. Hence, while a subordinate might enjoy unjustifiably high rewards, this will not increase this subordinate’s trust in the superior. The simulation results indicate that intra-organizational interdependencies in conjunction with the superior’s propensity to trust subtly affect the performance levels and trust-building on both the superior’s and subordinate’s sides.

References:

Brower, H. H., Schoorman, F. D., & Tan, H. H. (2000). A model of relational leadership: The integration of trust and leader–member exchange. The Leadership Quarterly, 11(2), 227-250. doi:https://doi.org/10.1016/S1048-9843(00)00040-0 Christ, M. H., Sedatole, K. L., & Towry, K. L. (2012). Sticks and Carrots: The Effect of Contract Frame on Effort in Incomplete Contracts. The Accounting Review, 87(6), 1913-1938. doi:10.2308/accr-50219 Dewatripont, M., Jewitt, I., & Tirole, J. (2000). Multitask agency problems: Focus and task clustering. European Economic Review, 44(4-6), 869-877. Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. The Academy of Management Review, 14(1), 57-74. Fehr, E., & Gächter, S. (2000). Fairness and Retaliation: The Economics of Reciprocity. Journal of Economic Perspectives, 14(3), 159-181. doi:10.1257/jep.14.3.159 Feltham, G. A., & Xie, J. (1994). Performance measure congruity and diversity in multi-task principal/agent relations. Accounting Review, 69(3), 429-453. Fisher, J. G., Maines, L. A., Peffer, S. A., & Sprinkle, G. B. (2005). An experimental investigation of employer discretion in employee performance evaluation and compensation. The Accounting Review, 80(2), 563-583. Hart, O. D. (1988). Incomplete Contracts and the Theory of the Firm. Journal of Law, Economics, & Organization, 4(1), 119-139. Hart, O. D. (2017). Incomplete Contracts and Control. American Economic Review, 107(7), 1731-1752. doi:10.1257/aer.107.7.1731 Holmström, B., & Milgrom, P. R. (2009). Multitask principal-agent analyses: incentive contracts, asset ownership, and job design In R. S. Kroszner & L. Putterman (Eds.), The Economic Nature of the Firm: A Reader (3 ed., pp. 232-244). New York: Cambridge University Press. Kauffman, S. A., & Levin, S. (1987). Towards a general theory of adaptive walks on rugged landscapes. Journal of Theoretical Biology, 128(1 (September)), 11-45. Kreps, D. M. (1990). Corporate culture and economic theory. Perspectives on positive political economy, 90, 8. Lambert, R. A. (2001). Contracting theory and accounting. Journal of Accounting and Economics, 32(1-3), 3-87. Sanga, S. (2018). Incomplete Contracts: An Empirical Approach. The Journal of Law, Economics, and Organization, 34(4), 650-679. doi:10.1093/jleo/ewy012 Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of organizational trust: Past, present, and future (Vol. 32, pp. 344-354): Academy of Management Briarcliff Manor, NY 10510. Sethi, R., & Somanathan, E. (2003). Understanding reciprocity. Journal of Economic Behavior & Organization, 50(1), 1-27. doi:https://doi.org/10.1016/S0167-2681(02)00032-X Tirole, J. (1999). Incomplete Contracts: Where do We Stand? Econometrica, 67(4), 741-781. doi:https://doi.org/10.1111/1468-0262.00052 Wall, F. (2022). Incomplete incentive contracts in complex task environments: an agent-based simulation with minimal intelligence agents. Journal of Economic Interaction and Coordination. doi:10.1007/s11403-022-00357-6

ABSTRACT. In this study we investigated the effects of borrower-based macroprudential policies on the Hungarian residential housing market. We quantified the results of ten different policy scenarios ranging from a regulation free situation to policies which can be considered outstandingly strict compared to the actual limits in any country. To carried out this analysis using an agent-based model which has several unique properties in the literature. It features a 1:1 scale mapping of the housing market, which makes it possi-ble to generate highly disaggregated results based on the income distribution of the households, the type of the transactions considered (i.e. purchases of first-time buyer households, home owner house-holds and investors) and the different geographic regions of the country. Furthermore, it includes all the relevant parts of the economy which are necessary for the realistic assessment of borrower-based poli-cies: households, a representative investor, a bank, and a construction company, which entities interact on the housing, rental, and credit markets. We found that borrower-based limits (especially the LTV ratio) can effectively restrain the dynamics of house prices, transaction numbers and the volume of housing loan issuance, which reduces the proba-bility of the emergence of house price bubbles and consequently the build-up of systemic financial risks. Our analysis also revealed that the effect of a 10 percentage point LTV limit change is not linear. It is the highest compared to the situation without regulation, and it gets smaller towards the stricter direction. Considering the composition of the transactions on the housing market we found that borrower-based measures constrain first-time buyer households the most strongly. While their proportion among the buyers decreases after the tightening of the limits, the weight of the investors and the households who already own a flat increase. In the case of a stricter LTV regulation, the demand on the rental market gets higher, and consequently the vacancy rate drops. Despite the lower rental fees (due to the lower house prices) in this scenario, the returns on the rental market are altogether higher. We applied a novel approach compared to the previous literature by evaluating the credit constraints of households using the Credit Availability indicator. This measure also considers the necessary income of households to sustain a minimal consumption level as an additional limit. This approach revealed that the lowest three deciles of households cannot obtain a housing loan regardless of the borrower-based limits due to their insufficient income. Consequently, if one would like to increase the share of the below the median income households who are eligible for a housing loan by a borrower-based policy change, only the 4th and 5th deciles can be affected, and even for these groups only the complete elimination of the limits leads to a noticeable improvement, which in turn would cause the buildup of considerable macro-prudential risks. An easing of the LTV rule by ten percentage points increases only the 6th – 8th deciles’ share in lending, while it decreases the share of the richest group of households. Another new element of our analysis is the geographic disaggregation of the results. Considering this dimension, we found that the households in the capital are constrained most strongly almost exclusively by the LTV limit, which result is due to the higher incomes and house prices in that region. In contrast, the requirement of a sufficient income to ensure the minimum consumption level is much stronger in the rest of the country. There are similarly strong differences in the case of house price dynamics. Not only the house prices, but also their growth rate is almost two times higher in the capital after an LTV easing, therefore, the probability of the emergence of a housing bubble is much higher in Budapest than in other regions of Hungary. Lastly, our study also contributes to the literature by considering the consequences of borrower-based policies on the construction sector. Although the long-term impact of the regulation compared to the regulation free scenario is inconclusive in this regard, we could clearly show that looser limits lead to more newly built flats and therefore a larger improvement of the housing stock’ quality than in the case of the stricter scenarios. Furthermore, our analysis also covers the renovation of flats. We found that a looser DSTI can have a moderate influence in this regard, as in the case poorer households who are una-ble to buy an own flat might be capable of taking out a loan for renovation at least. We tested the robustness of our findings using an alternative macroeconomic environment in which we assumed constant economic growth rate and permanently low interest rate to filter out the effects of cyclical patterns. Our main findings and the implications for policy makers remained intact in this sensitiv-ity analysis, furthermore, it provided us with valuable additional information. We found that even in this shock-free environment – if there is no macroprudential regulation – house price bubbles can build up endogenously. However, the emergence of this risk can be avoided if we apply even only loose borrow-er-based limits, and in this case financial stability concerns only arise in the model if there are also exog-enous macroeconomic shocks. A further finding of this exercise is the sensitivity of the impact of DSTI to the interest rate environment. The effect of a DSTI tightening decreases in this case due to the lower debt service, but it remains still considerably strong. After considering not only the financial stability aspect, but also the social implications and the conse-quences on the housing stock through the construction sector’s activity, we could compose a compre-hensive assessment of borrower-based macroprudential policies. We concluded that the tightening of either the LTV or the DSTI limit would not be beneficial from any of the three considered aspects com-pared to the current regulation. On the other hand, an extreme regulatory easing (i.e. the abolishing of the limits) would lead to a substantial increase in systemic financial stability risks while it would not im-prove the quality of the housing stock. The rest of the LTV-DSTI combinations all entail trade-offs be-tween the different dimensions of our evaluation framework, thus, the decision between them de-pends on the preferences of the policy makers. A looser LTV rule (90%) would improve the situation of low income and first-time buyer households, and it would also enhance the output of the construction sector, however these favourable consequences would come at the cost of greater financial stability risks due to the increased house price volatility and in some cases higher default rates. In summary, for policy makers with strong preferences, the 80% LTV is the more advantageous decision, while in the case of opting for higher economic activity and the decrease of inequality, the 90% LTV rule is the better choice. Supposing balanced preferences, the combination of a loose LTV and strict DSTI limit seem to be the best compromise. While the volatility of the house prices gets higher even in this policy scenario, it does not result in the escalation of defaults.

ABSTRACT. Several advanced and emerging economies experienced increasing house prices in the recent years, coupled with sig-nificantly lower income developments. First-time home buyers (FTBs) are particularly exposed to the current housing mar-ket conditions, because they finance their transactions by bank loans to a larger extent, and higher house prices propor-tionally to their income also entails higher amount of savings to secure their mortgage. Moreover, parallel to the rise of inflation, FTBs are also faced with higher interest rates, thus, an increasing share of FTBs become credit constrained and lose access to home ownership. This phenomenon is characterized by heterogeneity not only between countries, but also between domestic regions, e.g. housing affordability has typically deteriorated faster in capitals compared to the country-side in recent years. Supporting home ownership is one of the main goals of housing policy in many countries, thus the impact assessment of potential schemes supporting this group is highly relevant. We use an agent-based model representing the entirety of the Hungarian residential housing market consisting of nearly 10 million people, 4 million households and flats as well as all the housing loan contracts. The model features transactions in the housing and rental markets, a construction sector, buy-to-let investors, credit markets, house price dynamics, a procyclical banking sector regulated by a macroprudential authority and exogenous macro environment. All the relevant characteristics of households, flats and loan contracts are based on empirical micro-level data. The time span of each simulation was 12 years. We evaluated the effectiveness and compared the costs and benefits of the following 8 demand- and supply-side poli-cy schemes, both fiscal and macroprudential measures, available only for first-time home buyers: 1. Changing macroprudential regulation: the loan-to-value ratio (LTV) is increased to 90% and the debt service-to-income ratio (DSTI) is tightened by 10 percentage points. 2. Mortgage guarantee: LTV is increased to 90% and the state provides a guarantee up to 10% of the loan. 3. Percentage subsidy: 10% of the flat’s price which can be used by FTBs as down payment to borrow. FTBs belonging to the 10th income decile are not eligible and there is an upper limit on the flat’s value. 4. Lump-sum grant: 2.5 million HUF subsidy, which amount increases with inflation, and it can be used by FTB house-holds as down payment to borrow. In the first month the amount of the lump-sum grant is the same as the percentage subsidy for the average housing transaction. FTBs belonging to the 10th income decile are not eligible and there is an upper limit on the flat’s value. 5. Preferential state credit facility: The state provides a preferential loan up to 30% of the flat’s value with a 3% interest rate. Only 2% down payment is needed. The FTB does not have to pay installment for the first 5 years and the loan does not interest during this period. After 5 years, the installment increases with the debtor's income. FTBs belonging to the 10th income decile are not eligible. 6. Shared equity: The state owns 30% of the flat. Only 2% down payment is needed. For the first 5 years, the FTB does not have to pay anything, then it pays a rent (3% of the flat’s value). After 10 years, the FTB must buy back the state’s share at market price. FTBs belonging to the 10th decile are not eligible. 7. Rent to buy: The state buys the flat and the FTB must pay a monthly rent (75% of the market rent), which is consid-ered as capital repayment (with 0% interest rate). After 10 years, the FTB can pay the remaining capital and become the owner. While this scheme is very favorable, the eligibility criterium is strict: the participation is limited to only FTBs below median income. 8. Government as build for sale housing developer: The state builds new flats on state-owned sites in a non-profit way and sells these flats at lower than market price. The annual number of new flats built in this program is fixed. FTBs be-longing to the 10th income decile are not eligible. According to the results, 7 policy measures (except the build for sale) stimulates demand and increases house prices in the first 5 years. As a result, homeowner households will be better off during this period, as the value of their housing assets will also increase. However, 12 years after the introduction of the measures, the deviation of house prices from the base scenario is between -10% and +10%, so in case of some scenarios the value of the flat depreciates. In terms of com-position, every scheme shifts demand towards higher quality, increasing the number of newly built flats. The policy measures (apart from the build for sale) reduce both the utilization rate and the yields on the rental market, so the number of purchases for investment purposes is lower than in case of the baseline scenario. Each measure increases the overall number of FTBs, but there is a significant difference in their effectiveness. The lump-sum grant has the greatest effect, as it increases the number of FTBs by more than 40%. The equity share and the preferential state credit facility programs yield a 20% surplus of FTB transactions. In the case of the policy schemes of changing macroprudential regulation, mortgage guarantee, percentage subsidy, and rent to buy, the additional FTB-transaction effect is between 7 and 10%. However, the expansion is more even in the latter two cases, so the "shock" hits the housing market more evenly. The build for sale program has the smallest effect due to its design. We examined the effect of the 8 policies on the credit availability that is calculated at the level of individual house-holds, and it examines whether a household, which wants to buy an own home, can afford an average flat corresponding to its region and income decile ("justifiable housing need”), and if not, what is the main constraint it faces. There are 5 possi-ble values: buying a flat (1) without credit, (2) with credit but without credit constraints or borrowing is constrained by (3) LTV limit, (4) DSTI limit, (5) per capita income limit, as household’s income is not enough for minimum consumption and loan repayment. If the household is constrained by more limits, then it is categorized by the most binding limit. According to the credit availability indicator of FTBs calculated for the baseline scenario, income of households in the first two deciles is not sufficient to secure a loan, as they are constrained by the minimal consumption limit. Even in the 3-7th income deciles most of the households are credit constrained, mainly by the LTV, as their income is relatively low to accumulate the necessary savings. Most of the policies do not improve the creditworthiness of FTBs below the median income, as these households face other constraints. There are 3 exceptions, the lump-sum grant, the shared equity, and the preferential state credit facility, which ease credit constraints even in the 4th-5th income deciles. These policy schemes improve the creditworthiness of households to the greatest extent also in the 6th-9th deciles. Both the mortgage guarantee and the changing macroprudential regulation has a sizable positive effect on FTBs belonging to the 7th-10th income deciles. The effect of the percentage subsidy is similar, but without easing credit availability of FTBs in the 10th decile, as these households are not eligible for this program. The rent to buy and build for sale scenarios have almost no effect at all on credit availability. The overall evaluation of the policies is based on the comparison of their effectivity and tradeoffs (number of FTBs, fis-cal cost, borrowing capacity of lower income FTBs, renewal of housing stock, default rate, house price volatility): • The shared equity significantly improves credit availability and facilitates housing access for FTBs, it has a large positive effect on the quality of the housing stock, but it also increases financial stability risks, mainly due to the higher volatili-ty of house prices (which is twice as large as in the case of the baseline scenario). • The changing macroprudential regulation generates more FTB transactions through easing credit constraints for FTBs belonging to the 7-10th income deciles, it supports new construction, all this without fiscal cost, but these advantages come along with higher house price volatility. • The preferential state credit facility has no advantage compared to the shared equity (in terms of default rate, quality of housing stock and fiscal cost), while the mortgage guarantee is inferior to changing macroprudential regulation mainly due to the higher default rate. • The percent subsidy can increase the number of FTBs and the proportion of lower-income households in lending and facilitate the renewal of the housing stock with moderate house price volatility. However, it comes at a fiscal cost, and it is the least effective measure, as many FTBs benefit from it even though they would be able to buy a flat anyway. • The lump-sum grant increases the number of FTBs to the greatest extent and significantly eases the credit availability of medium income FTBs, but in the meantime it raises financial stability risks in terms of PD and LGD in the country-side, and it places a significant burden on the budget. • The two most targeted measures are the build for sale and the rent to buy schemes. The build for sale program has relatively moderate, but positive effects. The rent to buy scheme increases the number of FTBs by 10% without easing credit constraints and it supports new constructions. These targeted policies barely change financial risks, and their only disadvantage is the related fiscal cost.

ABSTRACT. World food markets are susceptible to systemic shocks due to a concentration of production in a few main breadbasket regions, strong import dependencies of many developing countries, and frequent unilateral and uncoordinated policy interventions such as export restrictions. The well-established agricultural integrated assessment model (IAMs), may assess longer-term food security risks arising from socioeconomic and climatic conditions, but they lack appraising the short-term global food system responses to weather extremes and policy-induced food supply failures. We here present the novel dynamic agent-based network model Agrimate which resolves the short-term trade and price dynamics at global agricultural markets with high temporal resolution. Driven by production anomalies and accounting for unilateral trade policy measures such as export restrictions, the model is designed to describe out-of-equilibrium market dynamics in crisis situations. We employ the model to wheat as a major food grain and hindcast the two recent world food price crises of 2007/08 and 2010/11. By disentangling the role of production failures and export restrictions, we contribute to the ongoing debate on the main short-term drivers of these crises. Next to the elaboration of regional hotspots of food insecurity for the present climate and socioeconomic structure, Agrimate will be also able to assess food security risks for future projected climate and socioeconomic conditions based on existing results of IAMs, which enhances the ability to evaluate environmental policy measures.

ABSTRACT. Discussions on how to complete Economic and Monetary Union (EMU) and increase its economic resilience have been going on since its germinal stage. In recent years, the Euro Debt Crisis first and, more recently, the Covid-19 pandemic have given a new impetus to such a debate and to the process of institutional transformation of the European Union. As already pointed out by Monnet (1978), each crisis indeed exerts a fundamental influence on the direction of the European integration process forcing the hand of reluctant policy actors. The Euro Debt Crisis marked a significant turning point for the monetary policy characterized by a far more active role played by the European Central Bank, inspired by a broadened interpretation of its mandate. The reform of European fiscal rules and the process of fiscal integration have instead lagged behind until the outbreak Covid-19 crisis which led to an unprecedented response by the European institutions. While the prompt launch of the massive pandemic emergency purchase programme (PEPP) by the ECB already marked a striking difference with the inertia affecting the ECB intervention in the aftermath of the Euro Crisis, the discontinuity with the past was even more striking on the fiscal side. In May 2020 the European Union agreed to suspend for the first time the stringent fiscal rules contained in the Stability and Growth Pact (SGP) to give more fiscal space to counteract the recession to those countries most hit by the coronavirus, even to those that were already highly indebted. During the same month the governments of France and Germany proposed a 500 billion Recovery Fund that paved the way to agreement of the European Council on the Next Generation EU (NGEU) Program in July 2020. The program was geared around the Recovery and Resilience Facility, a 750 billion fund, 390 billion to be distributed in grants and 360 in loans, aiming at funding reforms and investments in Member States up to the end of 2026. Besides the magnitude of the fiscal stimulus, the main novelty brought by the Franco-German original proposal and by the NGEU program lied in that they both broke two long-lasting European fiscal taboos. First, they legitimized the principle of fiscal transfers within the Union and the need to complement the rules designed to promote countries' fiscal discipline with a distinct redistributive dimension. Secondly, the deal called for the European Commission to borrow on the capital markets on behalf of the entire EU, thus opening the way to the issuance of common debt. It goes without saying that the symmetric and exogenous nature of the Covid-19 crisis gave a substantial contribution to the political acceptability of the proposal among those countries that have been traditionally more reluctant to accept, when not openly hostile, to a fiscal transfer union with shared debt. However, it is widely recognized that despite its exceptional and temporary character, the agreement can represent a major jump-forward in the process of European integration towards the transformation of the European Monetary Union into a fully-fledged Fiscal Transfer Union endowed with its own fiscal capacity. Exceptional policy measures originally introduced as temporary tools to face an emergency can indeed often become difficult to reverse later on (Derman and Verdun, 2018) The paper thus aims at providing a tentative assessment of the economic effects of transforming the European Union Monetary Union into an Intergovernamental Fiscal Transfer Union (IFTU) with its own fiscal capacity, partly funded by national contributions by member countries and partly by the issuance of common debt. We refer to a IFTU as a system in which national governments agree to share financial resources to address economic imbalances and promote stability, involving the transfer of funds from stronger member states to weaker ones, with the goal of supporting economic growth and stability across the Union. More precisely, we propose a mechanism of redistribution where each country contributes to a Union budget proportionally to its GDP and receives a share that is inversely related to the difference between its cyclical growth and the average of Union. Our analysis is carried out using the multi-country AB-SFC model of a Monetary Union presented in Caiani (2018, 2020). The model has been shown to yield reasonable values for the dynamics and relative dimension of key variables, broadly comparable with historical data and available stylized facts for the European Monetary Union in Caiani et al. (2018). More precisely, Caiani et al. (2018) investigated the effect of changes to the deficit-to-GDP threshold imposed to member countries by a Stability and Growth Pact-like legislation and how the dimension of the common market influence the results. In contrast, Caiani et al. (2020) examined the impact of alternative wage growth patterns on the economic dynamics of the Monetary Union, differentiating between changes occurring in a single country and coordinated changes that occur simultaneously across all countries.

In the present work, the model framework has been integrated by adding two novel institutional architectures: an IFTU funded by national contributions only, and a fully-fledged Fiscal Transfer Union (FFFTU) which foresees the possibility to issue common debt instruments. Results show that the IFTU and, even more, the FFFTU enhance the stability and performance of the Union GDP, without causing an increase of the overall debt burden. This result is achieved mainly through the stabilizing effect on international trade exerted by the redistribution rule which, providing support to the demand of countries with a worse cyclical performance, indirectly stabilizes their demand for the exports of other countries. In this way, a critical contagion channel through trade is softened. Our simulations also show that high-productivity, high-income `core' countries tends to benefit more from the introduction of a FFFTU thanks to their higher competitiveness on the internationally integrated market for tradables. This translates into a better dynamics of their external balance which eventually results into an improvement of their public debt-to-GDP ratios. This improvement makes `core' countries more likely to be net-contributors to the Union budget which in turn allows to improve in a balanced way the GDP growth also in the periphery. This result seems to be critical to evaluate the political acceptability of a FFFTU reverting the argument that the benefits of a fiscal transfer union would likely accrue primarily to the countries receiving transfers, while the costs would be spread among the wealthier ones, and suggesting instead that being net-contributors might be the consequence of the bigger benefits accrued by core countries. Finally, we tested the efficacy of the IFTU and FFFTU in tackling the consequences of exogenous and symmetric demand and supply shocks. Results suggests that a FFFTU seems to provide some beneficial effects when facing a demand shock, reducing the amplitude of the recession and the cumulative loss of GDP over time. Conversely, a Fiscal Transfer Union does not seem to bring significant advantages when faced with a shock to production. This result comes to no surprise given that the fiscal transfer union mainly operates as a stabilizer of aggregate demand.