Research

We set up a novel dataset for Italy combining i) balance sheet and income statement data on the universe of limited liability companies with ii) firm administrative records on market entry and exit and iii) the quasi-universe of public contracts. A procurement-firm puzzle arises: Holding the sector, region, or year fixed and controlling for size, age, and productivity, firms that receive public contracts survive longer. To identify the effect of public demand on firm survival, we rely on bid distribution to inform a regression discontinuity analysis. We find that the survival rate of winners relative to marginal losers is three p.p. (or 75%) higher after 36 months—two and a half years beyond the expiration of the median contract. We argue that this effect is long-lasting and goes beyond an earnings boost—a higher share of revenues from public agencies is a mechanism at stake. Public demand does not affect the productivity dynamics of survivors; it makes their future earnings increasingly dependent on public contracts and raises their financial leverage.

Media Coverage (IT - lavoce.info): Appalti che tengono in vita le imprese

This paper investigates the relationship between online misinformation, conspiracy theories, and fake news on Twitter during the COVID-19 pandemic in Italy. We examine the spread of COVID skepticism across Italian municipalities in relation to pre-COVID online fake news attitudes and other local socio-economic factors. We also analyze the impact of economic lockdowns on the spread of COVID skepticism and its relationship with vaccine hesitancy. Our findings indicate that strict policies in municipalities less affected by the pandemic led to a significant increase in online COVID skepticism. We also show that the spread of online COVID fake news influenced vaccine hesitancy, highlighting the potential unintended consequences of stringent and controversial policies on public opinion and health-related decisions. While legal coercion may temporarily address vaccine hesitancy, managing information biases can help individuals make better informed choices for their own and the public’s benefit.

In this paper, we provide an assessment of the evolution of markups in Italy in the last twenty years. To this aim, we resort to both macro and micro data and estimation techniques, namely reduced forms accounting measures (price-cost margins) and production function model-based indicators. When using aggregate data, we adopt a comparative approach and analyze markup dynamics in the four main euro area countries, whereas the micro-level analysis is focused on Italy. According to our findings i) markups have shown flat/slightly decreasing dynamics across EU countries, settling on average in level at 1.1;  ii) aggregate dynamics hide substantial across sector and firms heterogeneity; iii) the micro-level analysis for Italy indicates the within-firms component as the most relevant in explaining markups behavior; iv) no top firms-driven dynamics emerge. Our results results differ from those obtained by De Loecker and Eeckhout (2018) for Italy for two main reasons: first, our sample is more representative of the Italian corporate sectors as it includes non-listed firms; second, rather than  assuming common technology across countries, we estimate country specific technology parameters. Finally, we propose an encompassing measure of market power, summarizing the previously investigated indices in a principal component framework. We confirm its effectiveness based on a set of validation variables.

We estimate the effects of shop opening hours deregulation on the market structure of the retail sector and on the size and composition of the labor force employed in it. To identify the effects of interest, we exploit the staggered implementation of a reform that allowed Italian municipalities to adopt fully flexible opening hours in the late 1990s. Our findings indicate that the possibility of opening shops 24/7 increased employment in the retail sector by about 3% and raised the number of shops in the affected municipalities by about 2%. The effects were concentrated amongst workers employed in larger plants which were better able to exploit the flexibility introduced by the new regime. An analysis of individual-level evidence suggests that the deregulation also produced a recompostion of employment towards employees against self-employed.

Media Coverage: (IT) IlSole24Ore

The design of an effective system for contract procurement is a daunting task due to the implicit trade-off that suppliers’ competition creates between low awarding prices and poor contract enforcement. This paper analyses a key, but understudied institution that helps balance the supply risk in the context of the US federal procurement: the performance bonding. By adding a third party which guarantees the contract fulfillment between the supplier and the buyer, the performance bonding impacts both bidder selection and contract enforcement. We focus on the universe of federal contracts in the period 2005-2015, and exploit an exogenous variation in the threshold for the application of performance bonding and public oversight - i.e., the monitoring exerted by the public buyer itself - to separately estimate their causal effects on delays and cost overruns. We find that performance bonding positively affects outcomes, mostly through supplier selection. On the other hand, public oversight has a negative effect, in particular for lower competence buyers.

Related Awards: SIEP Prize 2017. Best unpublished paper by young economists - XXIX SIEP Conference. 

Media Coverage (IT - lavoce.info): Opere pubbliche in cerca di garanzie

This paper analyzes the impact of intermediaries’ concentration on the allocation of revenues in online platforms. We study sponsored search - the sale ad space on search engines through online auctions - documenting how advertisers increasingly bid through an handful of specialized intermediaries. This enhances automated bidding and data pooling, but lessens competition whenever the intermediary represents competing advertisers. Using data on nearly 40 million Google’s keyword-auctions, we first apply machine learning algorithms to cluster keywords into thematic groups serving as relevant markets. Then, through an instrumental variable strategy, we quantify a negative and sizeable impact of intermediaries’ concentration on platform’s revenues.

Media Coverage: (VoxEU) The rise of buyer power in online advertising; (ViaSarfatti25) Advertising Algorithms and Digital Oligopoly

We provide novel evidence on the effect of the threat of potential competition on the timing of entry in a new and growing industry. Exploiting a change in regulation in the Italian retail fuel market that generates exogenous variation in the number of potential entrants in the emerging Compressed Natural Gas segment, we show that markets with a higher number of potential entrants witness speedier entry decisions. We document that this result is likely driven by an increase in the incentives to preempt the market due to heightened risk of being anticipated by competitors.

Alongside instrumental-variables and fixed-effects approaches, the control function approach is the most widely used in production function estimation. Olley and Pakes (1996, Econometrica 64: 1263–1297), Levinsohn and Petrin (2003, Review of Economic Studies 70: 317–341), and Ackerberg, Caves, and Frazer (2015, Econometrica 83: 2411–2451) have all contributed to the field by proposing two-step estimation procedures, whereas Wooldridge (2009, Economics Letters 104: 112–114) showed how to perform a consistent estimation within a single-step generalized method of moments framework. In this article, we propose a new estimator based on Wooldridge's estimation procedure, using dynamic panel instruments `a la Blundell and Bond (1998, Journal of Econometrics 87: 115–143), and we evaluate its performance by using Monte Carlo simulations. We also present the new command prodest for production function estimation, and we show its main features and strengths in a comparative analysis with other community-contributed commands. Finally, we provide evidence of the numerical challenges faced when using the Olley–Pakes and Levinsohn–Petrin estimators with the Ackerberg–Caves–Frazer correction in empirical applications, and we document how the generalized method of moments estimates vary depending on the optimizer or starting points used. 

The Total Factor Productivity (TFP) - also called Multi-factor productivity - measuresthe change in output that cannot be accounted for by changes in the amounts of input.The R package prodest provides functions for TFP estimation following the most widely-known methodologies using the control function approach. Focusing on Value Addedproduction functions, it estimates the two–steps models presented by Olley–Pakes (1996)(Olley and Pakes 1996) and Levinshon–Petrin (2003) (Levinsohn and Petrin 2003), as wellas their correction proposed by Ackerberg–Caves–Frazer (2015) (Ackerberg, Caves, andFrazer 2015). The system GMM framework proposed by Wooldridge (2009) (Wooldridge2009) is also implemented in two slightly different versions. Dealing with standard Cobb-Douglas technology in a panel framework, all methods assume that the productivity termevolves according to a first-order Markov process and that a proxy variable exists - i.e., afunction of state variables and productivity - invertible and monotonically increasing inproductivity. Exploiting these features and with different choices of the proxy variables,the methods yield consistent estimates of labor and capital inputs parameters, allowingfor an immediate computation of TFP. The prodest package features also the Data Gen-erating Process used by Ackerberg–Caves–Frazer (2015) (Ackerberg, Caves, and Frazer2015) and allows for the simulation of datasets according to several measurement errorsand random shock variances. It can bet used by practitioners for both running MonteCarlo simulations and benchmarking estimate results.

The dynamics of markups both at the firm and at the economy-wide level has recently attracted renewed attention from scholars (De Loecker et al. 2020), institutions (Thum-Thysen et al. 2017) and broader audiences (The Economist, 2019). In this paper, I review the main methods for markup estimation both at the micro- and at the macro-level, and provide methodological insights as a guide for applied researchers. Finally, I present a new Stata module (markupest) which implements all methods, and an addition to the user-contributed module prodest, aimed at estimating markups after the estimation of production function. I show their main features and performance both on example firm-level production datasets and on National Accounts data, and further strengthen the results through a series of Monte Carlo simulations.

This paper quantifies the impact of online vaccine skepticism on pediatric vaccine uptake and health outcomes. We propose a novel methodology that combines Natural Language Processing and an instrumental variable strategy that leverages the intransitivity of the social network’s connections. By matching the universe of Italian vaccine-related tweets for 2013-2018 with vaccine coverage and preventable hospitalizations at the municipality level, we find that a 10pp increase in anti-vaccine sentiment causes i) a 0.43pp decrease in coverage of the Measles-Mumps-Rubella vaccine, ii) additional 2.1 hospitalizations among vulnerable populations per 100,000 residents, and iii) an 11% increase in the relevant healthcare expenses, equivalent to 7,311 euros. Drawing on the results of a simulated model, we further show the importance of targeted interventions to counter misinformation and improve vaccine uptake.

Media Coverage (VoxEU): The effects of vaccine scepticism on public health and the amplifying role of social media: Insights from Italy

Data are the key fuel of artificial intelligence and any change to the type and quality of available data has an impact on the type and performance of the feasible algorithms. We analyze the incentives that large digital platforms have to alter data flows to their advantage by strategically obfuscating data. We quantify this phenomenon in the context of digital advertising auctions through a series of simulated experiments where asymmetric bidders employ artificial intelligence algorithms to compete in Generalized second-price auctions. We find that when less detailed information is available to train algorithms, advertisers’ rewards are substantially and consistently lower.

This paper analyzes the process of centralization of public procurement in Europe, with emphasis on the Italian case. We analyze the normative and regulatory reforms occurring between 2000 and 2016 at both EU and Italian level, aimed at fostering purchasing centralization. We then empirically evaluate the potential distortions induced by the most recent wave of centralization reforms using procurement data. We find that administrations expecting to lose their ability to contract independently: i) in the short run, anticipate their purchases to avoid delegating to a central body, while, in the medium run, ii) manipulate contract values, breaking down purchases into smaller lots of amounts below the thresholds driving centralization requirements and iii) when given the option, aggregate into the smallest type of centralized purchasing bodies. These distortions partially offset the potential benefits of the centralization reform.

We use a quantitative model of wholesale electricity market in the spirit of Reguant (2019) to asses how different policy interventions affect electricity prices and quantities in Italy. To illustrate its possible applications,  we investigate three specific policies designed to address the energy crisis that unfolded in 2021-22:  (1) a EU-wide gas price cap, (2) a program of mandated demand reduction during peak hours, and (3) a national price cap on the gas used for electricity generation.

This paper studies whether and how much procurement managers matter for effective procurement outcomes. We rely on detailed data on Italian procurement for public works, and on the identity of public officials responsible for their tendering and execution. Our analysis shows that, ceteris paribus and even within the same procuring agency, their identity matters for effective procurement. Furthermore, for cleaner identification we exploit institutional changes that raised the stringency of eligibility criteria for such role as a lever of public officials’ quality and study how it affects procurement outcomes. In particular, we measure procurement success by the duration of the administrative procedures needed to award the tender and the time to complete it. We find that procurement managers’ quality positively relates to the speed of awarding and, thus, of completing the planned public works. 

We study the sales dynamics of grocery chain stores during the first wave of the COVID-19 pandemic in Italy. We document a sustained growth in revenues for storable products, such as food staples and household supplies, beginning right before restrictions on mobility were introduced, and lasting throughout the whole lockdown period. We also examine the revenue surge by disentangling the role of different types of stores. We find that the increase has been driven by the dynamics of smaller outlets, located in urban areas and closer to the city centre, while hypermarkets experienced a drop during the lockdown period, probably relating to their more peripheral position. We also exploit both the remarkable granularity of scanner data and the staggered implementation of restrictions across Italian regions to causally identify the short-term effects of mobility constraints on outlets' sales. According to our estimates, large grocery stores in areas subject to lockdown measures earned revenues around 10 per cent lower than their control group did.

Project duration is an important metric in the assessment of public works procurement and consists of the time taken to complete the three major sub-phases (design, awarding and execution). The overall duration may be influenced by various factors such as project characteristics, local market conditions, and the features of the contract ing authority. Italy is characterized by stark territorial differences, potentially encompassing all the above dimensions. This paper uses granular data on Italian procurement to investigate public works’ completion times in the last decade. We unveil performance differentials across macro-areas and analyse possible drivers. We find that i) Southern regions underperform with respect to those in the Centre-North, in particular in phases characterized by a greater intensity of administrative tasks; ii) durations are significantly correlated with the features of the contracting authority, such as workforce composition, workload and experience, and administrative efficiency; iii) these factors, however, explain the North-South divide only partially, suggesting the need to further analyse the internal functioning of contracting authorities.