景氣循環與結構變遷之兩部門模型分析

Abstract

本篇博士論文探討有關結構變遷與景氣循環之三項議題。第一項議題討論不同部門的總要素生產力成長與結構變遷在不同模型設定下的關係。本文認為農業生產力對於富國的長期且大幅的結構性變遷（農業就業比例隨時間逐漸萎縮）是關鍵的決定因素。本文進一步指出將農業消費的維生水準與農業資本納入標準的兩部門（農業與非農業）新古典成長模型中可以得到這個結果。當經濟成長時，農業消費的維生水準能促使經濟體系逐漸偏向非農業的消費。再者，本文發現，第一，資本與勞動的互補關係會產生勞動配置的延遲效果（hold-up effect），這個效果對於長期且大幅的結構變遷是很重要的，使得農業生產力所驅動的結構性變遷能與富國過去近兩百年的發展資料相一致。第二，農業生產力所驅動的結構性變遷比非農業生產力更能貼近富國過去兩個世紀的發展資料。 第二項議題剖析貨幣政策的衝擊對於耐久消費財與非耐久財之景氣循環的消費動態問題。根據實證文獻，經濟體系面對貨幣衝擊時，耐久消費財與非耐久財會同向變動，而且，耐久財價格相對於非耐久財價格更浮動。然而，標準的兩部門新古典模型（僵固的非耐久財價格與浮動的耐久財價格）在面對負向的貨幣衝擊時，兩者卻是反向變動（因為價格下降幅度不一致而產生替代效果，所以需求會偏向相對便宜的耐久財，而減少非耐久財的需求）。借貸摩擦（家計單位分成兩種：儲蓄者與面對抵押品信用限制的借款者）可以對借款者產生負的所得效果，進而減少兩種財貨的消費，但在耐久財價格為浮動下（耐久消費財之正的替代效果會大於負的所得效果），這個負的所得效果仍無法解決這個不一致問題。因此，本文擬將資本納入兩部門（耐久財與非耐久財）新古典模型（僵固的非耐久財價格與浮動的耐久財價格，以及借貸摩擦）以試圖解決這個不一致問題。模型機制如下：負向的貨幣衝擊會降低耐久財的相對價格，進而增加廠商對投資的需求（因為資本亦屬於耐久財部門），使得短期的投資成本增加，廠商的利潤減少。由於儲蓄者是廠商的股東，所以儲蓄者獲得的股利減少，進而降低儲蓄者的可支配所得，使得儲蓄者無法再增加太多的耐久消費財需求。因此，額外增加這個負的所得效果使得耐久消費財的總合需求（儲蓄者與借款者）減少，並與非耐久財同向波動，進而和實證結果一致。 近來有許多文獻證明投資衝擊是驅動景氣循環的最重要因子。然而，標準的實質景氣循環模型在面對正向的投資衝擊時，投資、產出與工時都增加，而消費卻減少（跨期替代效果：當期消費減少，而投資與未來消費增加），此與實證的景氣循環結果不一致。因此，生成本篇論文的第三項議題。為了解決這個景氣循環波動的問題，本文建構一個兩部門（消費財與投資財部門）新古典價格僵固模型，並依照景氣循環文獻將耐久消費財納入投資財中。在兩部門模型下，正向的投資衝擊會提高投資財（包含耐久財）相對於消費財（非耐久財）的價格，而較高的投資財（包含耐久財）價格會促使家計單位增加消費財（非耐久財）的需求。這個當期替代效果（消費增加）會大於跨期替代效果（消費減少）。因此，消費增加，與其它的實質總體變數同向波動，而符合實證的景氣循環結果。

This dissertation consists of three essays concerning structural change and business cycles in response to real and nominal shocks in multi-sector growth models. In the first essay, I study and compare the relationship between sectoral total factor productivity and economic structural transformation in different models. Several papers showed that growth in agricultural productivity was essential for today’s richest countries to take off early. However, few articles noticed that growth in agricultural productivity is critical in governing long-term and large structural changes in richest countries of today. This essay studies a two-sector neoclassical growth model with subsistence agricultural consumption and agricultural capital, and shows that (i) hold-up effect which follows on complementarity between capital and labor is crucial; and (ii) growth in agricultural productivity plays a more important role than growth in non-agricultural productivity in determining long-term and large structural changes in richest countries of today. Next, the second essay analyzes the problem of consumption dynamics between nondurables and consumer durables in response to monetary policy shocks. Empirical results indicate that consumer durables and nondurable consumption tend to comove in response to monetary policy shocks, and consumer durable prices are more flexible than nondurable prices. However, in otherwise standard two-sector neoclassical sticky-price models with flexible durable prices, following a contractionary monetary policy shock, nondurables decrease but consumer durables increase (substitution effect). Credit constraints can generate a negative income effect for borrowers because of tighter credit constraints resulted from a contractionary monetary policy. However, if durable prices are flexible, substitution effect dominates income effect, so the model still fails to generate joint decline. This essay resolves the comovement problem by adding firms’ investment and financial friction into a model with flexible durable prices. When investment is chosen by firms, contractionary monetary policy shocks reduce the relative price of durables, which induces investment and decreases firms’ real profits in the short run. Due to fewer profits remitted from firms, savers have a lower disposable income and cannot increase expenditures on consumer durables as much as otherwise. As a consequence, together with the credit constraint channel, aggregate consumer durables decrease and comove with nondurables. Finally, the third essay investigates the business cycle comovement problem driven by investment shocks. Recent research based on sticky-price models suggests that investment shocks are the most important driver of business cycle fluctuations. Yet, investment shocks generate the comovement problem because there is an intertemporal substitution effect away from consumption and toward investment, which is inconsistent with empirical evidence that consumption tends to comove along with other real macro variables procyclically. This essay resolves the comovement problem by extending the standard neoclassical sticky-price model to a two-sector model with consumer durable services. When investment goods are used for either capital investment or consumer durable services, positive investment shocks also generate an intratemporal substitution effect away from consumer durable services toward consumption. As the intratemporal substitution effect dominates the intertemporal substitution effect, consumption increases, and thus the comovement problem in business cycles is resolved.