International Economic Review ?2022年第四期

?

?

——更多動(dòng)態(tài)，請(qǐng)持續(xù)關(guān)注gzh：理想主義的百年孤獨(dú)

?

?

1.DEEP EQUILIBRIUM NETS

深度均衡網(wǎng)絡(luò)

Marlon Azinovic,?Luca Gaegauf,?Simon Scheidegger

?https://doi.org/10.1111/iere.12575

We introduce deep equilibrium nets (DEQNs)—a deep learning-based method to compute approximate functional rational expectations equilibria of economic models featuring a significant amount of heterogeneity, uncertainty, and occasionally binding constraints. DEQNs are neural networks trained in an unsupervised fashion to satisfy all equilibrium conditions along simulated paths of the economy. Since DEQNs approximate the equilibrium functions directly, simulating the economy is computationally cheap, and training data can be generated at virtually zero cost. We demonstrate that DEQNs can accurately solve economically relevant models by applying them to two challenging life-cycle models and a Bewley-style model with aggregate?risk.

我們引入了深度均衡網(wǎng)絡(luò)(DEQNs)——一種基于深度學(xué)習(xí)的方法，用于計(jì)算具有大量異質(zhì)性、不確定性和偶爾約束的經(jīng)濟(jì)模型的近似功能理性預(yù)期均衡。deqn是一種以無監(jiān)督方式訓(xùn)練的神經(jīng)網(wǎng)絡(luò)，以滿足經(jīng)濟(jì)模擬路徑上的所有均衡條件。由于deqn直接近似平衡函數(shù)，因此模擬經(jīng)濟(jì)的計(jì)算成本很低，并且可以以幾乎為零的成本生成訓(xùn)練數(shù)據(jù)。通過將deqn應(yīng)用于兩個(gè)具有挑戰(zhàn)性的生命周期模型和一個(gè)具有總風(fēng)險(xiǎn)的Bewley-style模型，我們證明了deqn可以準(zhǔn)確地求解經(jīng)濟(jì)相關(guān)模型。

?

?

2.INSIDE MONEY, INVESTMENT, AND UNCONVENTIONAL MONETARY POLICY

內(nèi)部貨幣、投資和非常規(guī)貨幣政策

Lukas Altermatt

First published: 31 May 2022

?https://doi.org/10.1111/iere.12588

I develop a model where banks play a central role in monetary policy transmission. By credibly committing to repayment, banks can perform liquidity transformation. Illiquid assets may pay a liquidity premium because they allow banks to create liquid assets. The policy analysis discusses how the monetary authority can affect nominal rates and inflation when the fiscal authority follows nominal or real debt targets. A main result is that under a nominal debt target, the monetary authority is only able to increase inflation at the zero-lower bound by issuing money via lump-sum transfers, while doing so via bond purchases is?ineffective.

我開發(fā)了一個(gè)模型，其中銀行在貨幣政策傳導(dǎo)中發(fā)揮核心作用。通過可靠地承諾還款，銀行可以進(jìn)行流動(dòng)性轉(zhuǎn)換。非流動(dòng)性資產(chǎn)可能會(huì)支付流動(dòng)性溢價(jià)，因?yàn)樗鼈冊(cè)试S銀行創(chuàng)造流動(dòng)性資產(chǎn)。政策分析討論了當(dāng)財(cái)政當(dāng)局遵循名義或?qū)嶋H債務(wù)目標(biāo)時(shí)，貨幣當(dāng)局如何影響名義利率和通貨膨脹。一個(gè)主要結(jié)果是，在名義債務(wù)目標(biāo)下，貨幣當(dāng)局只能通過一次性轉(zhuǎn)移發(fā)行貨幣來提高通脹水平，而通過購買債券來實(shí)現(xiàn)這一目標(biāo)是無效的。

?

?

3.THE COST OF GREENING STIMULUS: A DYNAMIC ANALYSIS OF VEHICLE SCRAPPAGE PROGRAMS

綠色刺激的成本:汽車報(bào)廢計(jì)劃的動(dòng)態(tài)分析

Shanjun Li,?Youming Liu,?Chao Wei

First published: 09 June 2022

?https://doi.org/10.1111/iere.12590

This article investigates the potential trade-off between the stimulus and environmental objectives in the context of the U.S. vehicle scrappage program. We develop and estimate a dynamic model of vehicle ownership and conduct a counterfactual analysis comparing the implemented policy with alternative designs. We find the cost of the implemented policy 25% higher in terms of induced sales and 64% higher in terms of induced spending than a policy design without the environmental objective. The findings serve as a caution for green stimulus programs to address both climate change and the economic crisis from the ongoing pandemic.

本文探討了在美國汽車報(bào)廢計(jì)劃的背景下，刺激計(jì)劃和環(huán)境目標(biāo)之間的潛在權(quán)衡。我們開發(fā)并估計(jì)了車輛擁有量的動(dòng)態(tài)模型，并進(jìn)行了反事實(shí)分析，將實(shí)施的政策與其他設(shè)計(jì)進(jìn)行比較。我們發(fā)現(xiàn)，與沒有環(huán)境目標(biāo)的政策設(shè)計(jì)相比，實(shí)施政策的成本在誘導(dǎo)銷售方面高出25%，在誘導(dǎo)支出方面高出64%。這一發(fā)現(xiàn)對(duì)綠色刺激計(jì)劃提出了警告，以應(yīng)對(duì)氣候變化和當(dāng)前大流行造成的經(jīng)濟(jì)危機(jī)。

?

?

?

4.MORAL HAZARD IN REMOTE TEAMS

遠(yuǎn)程團(tuán)隊(duì)的道德風(fēng)險(xiǎn)

Emilio Bisetti,?Benjamin Tengelsen,?Ariel Zetlin-Jones

First published: 14 April 2022

?https://doi.org/10.1111/iere.12579

We reexamine the ability of teams to credibly self-impose group punishments and prevent free riding when individual inputs are unobservable. We formulate self-imposed group punishments as performance underreporting by the team. Although underreporting is not credible in a static game, we show that simple strategies can sustain underreporting in a repeated game, and that the threat of underreporting improves welfare only if team members' preferences between shirking and team output consumption are nonseparable. Our results suggest that self-assessments can replace increased managerial monitoring in remote work?environments.

我們重新審視了團(tuán)隊(duì)可信地自我施加集體懲罰的能力，并在個(gè)人投入不可觀察的情況下防止搭便車。我們制定了自我施加的集體懲罰，如團(tuán)隊(duì)低報(bào)績效。盡管少報(bào)在靜態(tài)博弈中是不可信的，但我們證明了簡單的策略可以在重復(fù)博弈中維持少報(bào)，并且只有當(dāng)團(tuán)隊(duì)成員在逃避和團(tuán)隊(duì)產(chǎn)出消費(fèi)之間的偏好不可分離時(shí)，少報(bào)的威脅才會(huì)提高福利。我們的研究結(jié)果表明，在遠(yuǎn)程工作環(huán)境中，自我評(píng)估可以取代增加的管理監(jiān)控。

?

?

5.APPROXIMATE BAYESIAN INFERENCE AND FORECASTING IN HUGE-DIMENSIONAL MULTICOUNTRY VARs

大維多國變量的近似貝葉斯推理與預(yù)測

Martin Feldkircher,?Florian Huber,?Gary Koop,?Michael Pfarrhofer

First published: 30 March 2022

?https://doi.org/10.1111/iere.12577

Panel vector autoregressions (PVARs) are a popular tool for analyzing multicountry data sets. However, the number of estimated parameters can be enormous, leading to computational and statistical issues. In this article, we develop fast Bayesian methods for estimating PVARs using integrated rotated Gaussian approximations. We exploit the fact that domestic information is often more important than international information and group the coefficients accordingly. Fast approximations are used to estimate the latter whereas the former are estimated with precision using Markov chain Monte Carlo techniques. We illustrate, using a huge model of the world economy, that it produces competitive forecasts?quickly.

面板向量自回歸(PVARs)是分析多國數(shù)據(jù)集的常用工具。然而，估計(jì)參數(shù)的數(shù)量可能是巨大的，導(dǎo)致計(jì)算和統(tǒng)計(jì)問題。在本文中，我們開發(fā)了使用積分旋轉(zhuǎn)高斯近似估計(jì)PVARs的快速貝葉斯方法。我們利用國內(nèi)信息往往比國際信息更重要這一事實(shí)，并相應(yīng)地對(duì)系數(shù)進(jìn)行分組。后者采用快速近似估計(jì)，而前者則采用馬爾可夫鏈蒙特卡羅技術(shù)進(jìn)行精確估計(jì)。我們用一個(gè)巨大的世界經(jīng)濟(jì)模型來說明，它能迅速產(chǎn)生有競爭力的預(yù)測。

?

6.POSITIVE SKILL CLUSTERING IN ROLE-ASSIGNMENT MATCHING MODELS

Axel Anderson

First published: 25 March 2022

?https://doi.org/10.1111/iere.12576

Kremer and Maskin (1996, NBER Working Paper) explore matching when production involves defined roles. Despite underlying Cobb–Douglas production functions, the induced maximum production function is not supermodular (SPM), and positive sorting does not arise. This article introduces and solves a general class of role-assignment matching models with general SPM production functions. The unique equilibrium entails a novel blend of positive sorting in the large, and locally negative sorting that I call positive clustering. I show how the equilibrium matching changes as the production function changes. In a dynamic extension, sorting, mobility, and wage inequality positively covary with changes in production across matching markets.

Kremer和Maskin (1996, NBER工作論文)探討了當(dāng)生產(chǎn)涉及定義角色時(shí)的匹配。盡管存在Cobb-Douglas產(chǎn)生函數(shù)，但誘導(dǎo)的最大產(chǎn)生函數(shù)不是超模的(SPM)，并且不會(huì)產(chǎn)生正排序。本文介紹并求解了一類具有通用SPM生產(chǎn)函數(shù)的角色分配匹配模型。這種獨(dú)特的平衡需要一種新的大的正排序和局部的負(fù)排序的混合，我稱之為正聚類。我展示了平衡匹配是如何隨著生產(chǎn)函數(shù)的變化而變化的。在動(dòng)態(tài)擴(kuò)展中，在匹配市場中，排序、流動(dòng)性和工資不平等與生產(chǎn)變化正相關(guān)。

?

?

7.ENDOGENOUS SOCIAL NETWORKS AND INEQUALITY IN AN INTERGENERATIONAL SETTING

內(nèi)源性社會(huì)網(wǎng)絡(luò)與代際環(huán)境中的不平等

Yannis M. Ioannides

First published: 20 April 2022

?https://doi.org/10.1111/iere.12578

When providing professional services, an expert may misbehave by either prescribing “wrong” treatment for consumer's problem or failing to exert proper effort to diagnose it. We show that under a range of liabilities the expert will recommend the appropriate treatment based on his private information if markups for alternative treatments are close enough; however, a well-designed liability rule is essential for also motivating efficient diagnosis effort. We further demonstrate that unfettered price competition between experts may undermine the efficient role of liability, whereas either a minimum-price constraint or an obligation-to-serve requirement can restore?it.

在提供專業(yè)服務(wù)時(shí)，專家可能會(huì)表現(xiàn)不當(dāng)，要么對(duì)消費(fèi)者的問題開出“錯(cuò)誤”的治療方案，要么沒有做出適當(dāng)?shù)脑\斷。我們表明，在一定的責(zé)任范圍下，如果替代治療的加成足夠接近，專家將根據(jù)他的私人信息推薦適當(dāng)?shù)闹委煼椒?然而，設(shè)計(jì)良好的責(zé)任規(guī)則對(duì)于激勵(lì)有效的診斷工作也是必不可少的。我們進(jìn)一步證明，專家之間不受約束的價(jià)格競爭可能會(huì)破壞責(zé)任的有效作用，而最低價(jià)格限制或服務(wù)義務(wù)要求都可以恢復(fù)責(zé)任的有效作用。

?

?

?

8.THE NEUTRALITY OF NOMINAL RATES: HOW LONG IS THE LONG RUN?

名義利率的中立性:長期有多長?

Jo?o Valle e Azevedo,?Jo?o Ritto,?Pedro Teles

First published: 18 May 2022

?https://doi.org/10.1111/iere.12584

We revisit the empirical relationship between policy interest rates and inflation aimed at understanding the reasons for persistently low inflation in Japan and the euro area and why monetary policy has been unable to raise it. We document the long-run positive relationship between nominal rates and inflation across countries and across time. We also find that a permanent rise in nominal rates leads to an increase in inflation, also in the short run. These findings suggest that the low inflation outcomes are a result of policy rates being kept too low for too?long.

我們回顧了政策利率和通貨膨脹之間的經(jīng)驗(yàn)關(guān)系，旨在理解日本和歐元區(qū)通貨膨脹持續(xù)低的原因，以及為什么貨幣政策一直無法提高通貨膨脹率。我們記錄了不同國家和不同時(shí)間名義利率和通貨膨脹之間的長期正關(guān)系。我們還發(fā)現(xiàn)，名義利率的長期上升會(huì)導(dǎo)致通脹的上升，也是在短期內(nèi)。這些發(fā)現(xiàn)表明，低通脹的結(jié)果是政策利率維持在過低水平太長時(shí)間的結(jié)果。

?

?

9.HETEROGENEOUS TREATMENT EFFECTS OF NUDGE AND REBATE: CAUSAL MACHINE LEARNING IN A FIELD EXPERIMENT ON ELECTRICITY CONSERVATION

助推和返利的異質(zhì)處理效應(yīng):節(jié)電現(xiàn)場實(shí)驗(yàn)中的因果機(jī)器學(xué)習(xí)

Kayo Murakami,?Hideki Shimada,?Yoshiaki Ushifusa,?Takanori Ida

First published: 03 June 2022

?https://doi.org/10.1111/iere.12589

This study investigates the different impacts of monetary and nonmonetary incentives on energy-saving behaviors using a field experiment conducted in Japan. We find that the average reduction in electricity consumption from the rebate is 4%, whereas that from the nudge is not significantly different from zero. Applying a novel machine learning method for causal inference (causal forest) to estimate heterogeneous treatment effects at the household level, we demonstrate that the nudge intervention's treatment effects generate greater heterogeneity among households. These findings suggest that selective targeting for treatment increases the policy efficiency of monetary and nonmonetary interventions.

本研究通過在日本進(jìn)行的現(xiàn)場實(shí)驗(yàn)，探討了貨幣激勵(lì)和非貨幣激勵(lì)對(duì)節(jié)能行為的不同影響。我們發(fā)現(xiàn)，從退稅中平均減少了4%的用電量，而從輕推中減少的用電量與零沒有顯著差異。應(yīng)用一種新穎的因果推斷(因果森林)機(jī)器學(xué)習(xí)方法來估計(jì)家庭層面的異質(zhì)治療效果，我們證明了助推干預(yù)的治療效果在家庭之間產(chǎn)生了更大的異質(zhì)性。這些發(fā)現(xiàn)表明，選擇性靶向治療可以提高貨幣和非貨幣干預(yù)的政策效率。

?

10.NETWORK EXTERNALITIES, DOMINANT VALUE MARGINS, AND EQUILIBRIUM UNIQUENESS

網(wǎng)絡(luò)外部性、支配價(jià)值邊際和均衡唯一性

JAY PIL CHOI,?CHRISTODOULOS STEFANADIS

First published: 25 May 2022

?https://doi.org/10.1111/iere.12587

We examine tippy network markets that accommodate price discrimination. The analysis shows that when a mild equilibrium refinement, the monotonicity criterion, is adopted, network competition may have a unique subgame-perfect equilibrium regarding the winner's identity; the prevailing brand may be fully determined by its product features. We bring out the concept of the dominant value margin, which is a metric of the effectiveness of divide-and-conquer strategies. The supplier with the larger dominant value margin may always sell to all customers in equilibrium. Such a market outcome is not necessarily socially efficient.

我們研究了適應(yīng)價(jià)格歧視的網(wǎng)絡(luò)市場。分析表明，當(dāng)采用溫和的均衡細(xì)化，即單調(diào)性準(zhǔn)則時(shí)，網(wǎng)絡(luò)競爭對(duì)于勝利者的身份可能具有唯一的子博弈-完美均衡;主導(dǎo)品牌可能完全由其產(chǎn)品特征決定。我們提出了優(yōu)勢(shì)價(jià)值邊際的概念，這是一個(gè)分治策略有效性的度量標(biāo)準(zhǔn)。具有較大優(yōu)勢(shì)價(jià)值邊際的供應(yīng)商可能總是均衡地銷售給所有客戶。這樣的市場結(jié)果并不一定對(duì)社會(huì)有效。

?

?

?

11.LEARNING ABOUT REGIME CHANGE

了解政權(quán)更迭

Andrew Foerster,?Christian Matthes

First published: 18 May 2022

?https://doi.org/10.1111/iere.12585

Total factor productivity (TFP) and investment specific technology (IST) growth both exhibit regime-switching behavior, but the regime at any given time is difficult to infer. We build a rational expectations real business cycle model where the underlying TFP and IST regimes are unobserved. We develop a general perturbation solution algorithm for a wide class of models with unobserved regime-switching. Using our method, we show learning about regime-switching fits the data, affect the responses to regime shifts and intraregime shocks, increase asymmetries in the responses, generate forecast error bias even with rational agents, and raise the welfare cost of?fluctuations.

全要素生產(chǎn)率(TFP)和投資專用性技術(shù)(IST)增長均表現(xiàn)出區(qū)制轉(zhuǎn)換行為，但在任何給定時(shí)間的區(qū)制都難以推斷。我們建立了一個(gè)理性預(yù)期的實(shí)際經(jīng)濟(jì)周期模型，其中基本的TFP和IST制度是不可觀測的。針對(duì)一類具有不可觀測區(qū)域切換的模型，我們開發(fā)了一個(gè)通用的攝動(dòng)解算法。利用我們的方法，我們發(fā)現(xiàn)，對(duì)區(qū)制轉(zhuǎn)換的學(xué)習(xí)擬合了數(shù)據(jù)，影響了對(duì)區(qū)制轉(zhuǎn)換和區(qū)制內(nèi)部沖擊的反應(yīng)，增加了反應(yīng)的不對(duì)稱性，即使在理性行為下也會(huì)產(chǎn)生預(yù)測誤差偏差，并提高了波動(dòng)的福利成本。

?

?

——更多動(dòng)態(tài)，請(qǐng)持續(xù)關(guān)注gzh：理想主義的百年孤獨(dú)