影響營建機具價格因素之探討-以美國為例 - 政大學術集成

全文

(1)國立政治大學應用經濟與社會發展英語碩士學位學程 International Master’s Program of Applied Economics and Social Development College of Social Sciences National Chengchi University. 立. 政治大. ‧ 國. 學. 影響營建機具價格因素之探討-以美國為例. ‧. Determinants of Construction Equipment Prices in the United States er. io. sit. y. Nat. Anthony J. Laurence. n. Dr. Tsoyu Calvin Lin aAdvisor: iv l C 2020年3月 n hengchi U March 2020 National Chengchi University. 碩士論文 Master’s Thesis. DOI:10.6814/NCCU202000410.

(2) Determinants of Construction Prices in the 政治Equipment 大立 United States. ‧. ‧ 國. 學 Anthony J. Laurence. n. al. er. io. sit. y. Nat. March 2020. Ch. engchi. i n U. v. DOI:10.6814/NCCU202000410.

(3) A Thesis Submitted to International Master’s Program of Applied Economics and Social Development National Chengchi University. 立. March 2020 治政大 Abstract. ‧ 國. 學. Equipment production prices have been shown to be a reflection of the overall health of economies. Additionally, this is also true for Construction Equipment Prices (CEPs) in addition to. ‧. construction-related equipment prices such as excavator, crane, and dragline prices (ECDPs), and general machinery prices (GMPs). On top of this, technological progress over the past few decades. Nat. sit. y. has lowered prices due to increased efficiency for many asset classes including capital. However,. io. er. correlates of equipment prices in general has been largely ignored in the academic literature. This is even more so true for CEPs, ECDPs, and GMPs. What then are the correlates of CEPs, ECDPs,. n. al. Ch. i n U. v. and GMPs? Using monthly data from 1993 until 2019 from the Bank of the Federal Reserve in St.. engchi. Louis, this thesis uses ordinary least square (OLS) regressions and vector autoregressions (VARs) the latter of which implements orthogonalized impulse response functions (OIRFs) to determine what the long-term correlates and in some instances short-term correlates of CEPs, ECDPs, and GMPs. The results of the OLS method and the VAR’s OIRFs show that a number of technological, macroeconomic, and commodity variables consistently correlate with these indices. However, the most consistently correlated variables are those associated with financial markets. Key Terms: Construction Equipment Prices; Excavator, Crane, and Dragline Prices; Ordinary Least Square Regression; Vector Autoregression; United States Construction Industry. 4. DOI:10.6814/NCCU202000410.

(4) Table of Contents 1.. Introduction ........................................................................................................................................ 13. 2.. Literature Review ................................................................................................................................ 17 2.1.. Interest Rates and Investment .................................................................................................... 21. 2.2.. Growth, Supply, Demand, Labor, and Commodities .................................................................. 24. 2.3.. Oligopolistic Competition in the US Construction Industry ........................................................ 26. 2.4.. Equipment Prices and Other Financial Market Indices ............................................................... 29. 2.5.. Machinery or Equipment Prices or Inflation and Technological Innovation .............................. 33. 2.6.. Tariff Impacts on CEPs................................................................................................................. 36. 2.8.. Methodology....................................................................................................................................... 42 Descriptive Statistics and Properties of Variables ...................................................................... 45. 3.2.. Ordinary Least Square Linear Regression Method ..................................................................... 52. 3.3.. Vector Auto-Regression Models ................................................................................................. 53. 3.4.. Augmented Dickey Fuller Unit Root Tests .................................................................................. 56. 3.5.. Lag Order Selection Criterion...................................................................................................... 62. 3.6.. Impulse Response Functions....................................................................................................... 63. 3.7.. Orthogonalized Impulse Response Functions............................................................................. 64. er. io. sit. y. Nat. al. iv n C Granger Causality ........................................................................................................................ 66 hengchi U Dynamic Forecasts ...................................................................................................................... 66 n. 3.9.. ‧. 3.1.. 3.8.. 3.10. 4.. 學. 3.. ‧ 國. 2.7.. 政治大 Price Versus Cost and Post-Keynesian Price Theory ................................................................... 36 立 Political Economy of the US Construction Industry .................................................................... 37. Added Variable Plots ............................................................................................................... 67. Empirical Results ................................................................................................................................. 69 4.1.. Ordinary Least Square Regressions............................................................................................. 69. 4.1.1.. CEPs Model 1—NASDAQ, Data Processing, and Technology CPI........................................ 74. 4.1.2.. CEPs Model 2— Diesel Prices, DXY, and Inventory Over Sales ........................................... 84. 4.1.3.. CEPs Model 3—GDP, Interest Rates, and Industrial Production ........................................ 94. 4.1.4.. CEPs Model 4—One-Month Differenced NASDAQ, Interest Rates, and Steel Prices ....... 104. 4.2.. ECDP VAR Models ..................................................................................................................... 112. 4.2.1.. ECDP Model 1—Data Processing, the NASDAQ, and Technology Hardware CPI ............. 114. 4.2.2.. ECDP Model 2—DXY, Interest Rates, and GDP ................................................................. 123 5. DOI:10.6814/NCCU202000410.

(5) 4.2.3. 4.3.. Analysis of the Results ...................................................................................................................... 156 5.1.. Bivariate Relationships of CEPs ................................................................................................. 156. 5.1.1.. The NASDAQ Composite and CEPs........................................................................................ 156. 5.1.2.. Macroeconomic Variables .................................................................................................... 172. 5.1.3.. NRCS ...................................................................................................................................... 179. 5.1.4.. Gross Domestic Product........................................................................................................ 181. 5.1.5.. Iron Ore, Steel, Diesel, and the DXY ...................................................................................... 181. 5.1.6.. Industrial Production of Business Equipment ....................................................................... 182. 政治大. 5.2.. General-Purpose Equipment Correlates ................................................................................... 183. 5.3.. Excavators, Cranes, and Dragline OLS Correlations .................................................................. 191. 5.4.. Vector Auto-Regression Orthogonalized Impulse Responses .................................................. 195. 5.5.. Summary and Ramification of Technical Analysis..................................................................... 200. 5.6.. Granger Causality Results ......................................................................................................... 206. 5.7.. Political Party Impacts ............................................................................................................... 207. 立. 學. ‧. Conclusion ......................................................................................................................................... 220. io. sit. y. Nat. n. al. er. 6.. Added Variable Plots ................................................................................................................. 142. ‧ 國. 5.. ECDP Model 3—Industrial Production, Iron and Steel Prices, and Diesel Prices .............. 134. Ch. engchi. i n U. v. 6. DOI:10.6814/NCCU202000410.

(6) LIST OF FIGURES FIGURE 2-1 THE BON CURVE. FROM BON (1992). ........................................................................................................ 18 FIGURE 2-2 OLIGOPOLISTIC PRICE MANIPULATION AND PROFIT TAKING. ................................................................... 27 FIGURE 2-3 HOW OLIGOPOLIES CAN INCREASE PRICE WHILE ALSO INCREASING SUPPLY .......................................... 28 FIGURE 2-4 GMPS AND PCS ONE-YEAR DIFFERENCED ............................................................................................... 39 FIGURE 2-5 NRCS AND PCS ONE-YEAR DIFFERENCED ............................................................................................... 40 FIGURE 2-6 PCS AND GDP ONE-YEAR DIFFERENCED ................................................................................................. 41 FIGURE 3-1 HISTOGRAM FOR MESOKURTIC CONSTRUCTION EQUIPMENT PRICES AND LEPTOKURTIC IRON AND STEEL PRICES ................................................................................................................................................................. 51 FIGURE 4-1 OIRF OF DATA PROCESSING IMPULSE AND CEPS RESPONSE .................................................................... 76 FIGURE 4-2 OIRF OF TECHNOLOGY HARDWARE CPI IMPULSE AND CEPS RESPONSE ................................................. 77 FIGURE 4-3 OIRF OF THE NASDAQ COMPOSITE AND CEPS RESPONSE ...................................................................... 78 FIGURE 4-4 OIRF OF CEPS IMPULSE AND CEPS RESPONSE ......................................................................................... 79 FIGURE 4-5 GRANGER CAUSALITY MAP FOR THE NASDAQ, CEPS, DATA PROCESSING, AND TECHNOLOGY CPI ...... 81 FIGURE 4-6 DYNAMIC FORECASTS FOR FOR NASDAQ, CEPS, DATA PROCESSING, AND TECHNOLOGY CPI ............... 83 FIGURE 4-7 OIRF OF DIESEL PRIES IMPULSE AND CEPS RESPONSE ............................................................................. 86 FIGURE 4-8 OIRF OF IRON AND STEEL PRIES ONE-MONTH DIFFERENCED IMPULSE AND CEPS ONE-YEAR DIFFERENCED RESPONSE ..................................................................................................................................... 87 FIGURE 4-9 OIRF OF THE DXY IMPULSE AND CEPS RESPONSE................................................................................... 89 FIGURE 4-10 GRANGER CAUSALITY MAP FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS......................................................................................................................................... 92 FIGURE 4-11 DYNAMIC FORECASTS FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS ............................................................................................................................................... 93 FIGURE 4-12 OIRF OF ONE-MONTH DIFFERENCED GDP IMPULSE AND CEPS RESPONSE ............................................ 96 FIGURE 4-13 OIRF OF ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES IMPULSE AND CEPS RESPONSE ............................................................................................................................................................................ 97 FIGURE 4-14 OIRF OF ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT IMPULSE AND CEPS RESPONSE .................................................................................................................................................. 98 FIGURE 4-15 GRANGER CAUSALITY MAP FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT ....................................................................................................................................................... 101 FIGURE 4-16 DYNAMIC FORECASTS FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT .......................................................................................................................................................................... 102 FIGURE 4-17 OIRF FOR ONE-MONTH DIFFERENCED NASDAQ IMPULSE AND ONE-YEAR DIFFERENCED CEPS RESPONSE .......................................................................................................................................................... 107 FIGURE 4-18 OIRF FOR ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES IMPULSE AND ONE-YEAR DIFFERENCED CEPS RESPONSE ......................................................................................................................... 108 FIGURE 4-19 OIRF FOR ONE-MONTH DIFFERENCED IRON AND STEEL PRICES IMPULSE AND ONE-YEAR DIFFERENCED CEPS RESPONSE ................................................................................................................................................ 109 FIGURE 4-20 GRANGER CAUSALITY MAP FOR ONE-MONTH DIFFERENCED NASDAQ COMPOSITE, ONE-MONTH DIFFERENCED IRON AND STEEL PRICES, AND ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES . 111 FIGURE 4-21 OIRF OF ONE-YEAR DIFFERENCED TECHNOLOGY HARDWARE CPI IMPULSE AND ECDPS RESPONSE . 116 FIGURE 4-22 OIRF OF ONE-YEAR DIFFERENCED NASDAQ COMPOSITE AND ECDPS RESPONSE ............................. 117 FIGURE 4-23 OIRF OF ONE-YEAR DIFFERENCED DATA PROCESSING CPI IMPULSE AND ECDPS RESPONSE ............. 118 FIGURE 4-24 OIRF OF ONE-YEAR DIFFERENCED ECDPS IMPULSE AND ECDPS RESPONSE ...................................... 119 FIGURE 4-25 DYNAMIC FORECAST RESULTS FOR ONE-YEAR DIFFERENCED ECDPS, THE NASDAQ COMPOSITE, DATA PROCESSING CPI, AND TECHNOLOGY HARDWARE CPI ........................................................................... 122 FIGURE 4-26 OIRF FOR THE DXY IMPULSE AND ECDPS RESPONSE.......................................................................... 126 FIGURE 4-27 OIRF FOR ONE-MONTH DIFFERENCED INTEREST RATES IMPULSE AND ECDPS RESPONSE .................. 128 FIGURE 4-28 OIRF FOR ONE-MONTH DIFFERENCED GDP IMPULSE AND ECDPS RESPONSE ..................................... 129. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 7. DOI:10.6814/NCCU202000410.

(7) FIGURE 4-29 GRANGER CAUSALITY MAP FOR ECDPS, ONE-YEAR DIFFERENCED DXY, ONE-MONTH DIFFERENCED INTEREST RATES, AND ONE-MONTH DIFFERENCED GDP .................................................................................. 132 FIGURE 4-30 DYNAMIC FORECASTS FOR ECDPS, ONE-YEAR DIFFERENCED DXY, ONE-MONTH DIFFERENCED INTEREST RATES, AND ONE-MONTH DIFFERENCED GDP .................................................................................. 133 FIGURE 4-31 OIRF FOR DIESEL PRICES IMPULSE AND ECDPS RESPONSE .................................................................. 136 FIGURE 4-32 OIRF FOR INDUSTRIAL PRODUCTION IMPULSE AND ECDS RESPONSE .................................................. 137 FIGURE 4-33 OIRF FOR IRON AND STEEL PRICES IMPULSE AND ECDS RESPONSE ..................................................... 138 FIGURE 4-34 DYNAMIC FORECASTS FOR ECDPS, IRON AND STEEL PRICES ONE-MONTH DIFFERENCED, INDUSTRIAL PRODUCTION ONE-MONTH DIFFERENCED, AND DIESEL PRICES ONE-YEAR DIFFERENCED............................... 141 FIGURE 4-35 ADDED VARIABLE PLOTS MATRIX FOR ONE-YEAR DIFFERENCED VARIABLES..................................... 143 FIGURE 4-36 IRF FOR HOUSE IMPULSE AND CEPS RESPONSE .................................................................................... 148 FIGURE 4-37 IRF FOR THE SENATE IMPULSE AND CEPS RESPONSE ........................................................................... 149 FIGURE 4-38 CEPS IMPULSE HOUSE CONTROL RESPONSE OIRF ............................................................................... 151 FIGURE 4-39 HOUSE POLITICAL CONTROL AND NRCS .............................................................................................. 153 FIGURE 4-40 NRCS IMPULSE PCS RESPONSE ............................................................................................................ 154 FIGURE 5-1 CAPACITY UTILIZATION 1-MONTH CHANGE AND UNEMPLOYMENT 1-MONTH CHANGE. FROM FEDERAL RESERVE BANK OF ST. LOUIS (2019)................................................................................................................. 158 FIGURE 5-2 CEPS ONE-YEAR DIFFERENCE AND THE NASDAQ COMPOSITE ONE-YEAR DIFFERENCE, 1993-2008 AND 2009-2019 ......................................................................................................................................................... 159 FIGURE 5-3 R&D ONE-YEAR DIFFERENCE AND THE NASDAQ COMPOSITE ONE-YEAR DIFFERENCE ...................... 163 FIGURE 5-4 CEPS ONE-YEAR DIFFERENCE AND TECH CPI ONE-YEAR DIFFERENCE, 1993-2009 .............................. 164 FIGURE 5-5 CEPS ONE-YEAR DIFFERENCE AND TECH CPI ONE-YEAR DIFFERENCE, 2009-2019 .............................. 166 FIGURE 5-6 NOMINAL MANUFACTURING EMPLOYEES AND NOMINAL INDUSTRIAL PRODUCTION 1978-2019 ........... 168 FIGURE 5-7 CEPS ONE-YEAR DIFFERENCED, ECDPS ONE-YEAR DIFFERENCED AND DATA PROCESSING PPI ONEYEAR DIFFERENCE ............................................................................................................................................ 171 FIGURE 5-8 CEPS ONE-YEAR DIFFERENCE AND R&D EMPLOYEES IN CALIFORNIA ONE-YEAR DIFFERENCE ........... 172 FIGURE 5-9 CEPS ONE-YEAR DIFFERENCE AND THE RATIO OF INVENTORY OVER SALES IN MANUFACTURING ONEYEAR DIFFERENCE ............................................................................................................................................ 174 FIGURE 5-10 INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT ONE-YEAR DIFFERENCE AND THE RATIO OF INVENTORY OVER SALES ONE-YEAR DIFFERENCE ........................................................................................... 176 FIGURE 5-11 CEPS ONE-YEAR DIFFERENCE AND THE EFFECTIVE FEDERAL FUND RATE ONE-YEAR DIFFERENCED . 178 FIGURE 5-12 CEPS ONE-YEAR DIFFERENCE AND NRCS ONE-YEAR DIFFERENCE .................................................... 180 FIGURE 5-13 CEPS ONE-YEAR DIFFERENCE AND INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT ONE-YEAR CHANGE ............................................................................................................................................................ 183 FIGURE 5-14 CEPS ONE-YEAR DIFFERENCE AND GMPS ONE-YEAR DIFFERENCE .................................................... 184 FIGURE 5-15 CEPS ONE-YEAR DIFFERENCE AND GMPS ONE-YEAR DIFFERENCE .................................................... 185 FIGURE 5-16 CEPS ONE-YEAR DIFFERENCED, GMPS PPI ONE-YEAR DIFFERENCED, AND THE RATIO OF INVENTORY OVER SALES ONE-YEAR DIFFERENCED ............................................................................................................. 190 FIGURE 5-17 CEPS, ECDPS, AND GMPS .................................................................................................................... 192 FIGURE 5-18 ECDPS ONE-YEAR DIFFERENCE AGAINST CEPS ONE-YEAR DIFFERENCE ........................................... 194 FIGURE 5-19 CEPS, ECDPS, AND GDP ...................................................................................................................... 197 FIGURE 5-20 GDP AND INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT ONE-YEAR DIFFERENCED ................... 198 FIGURE 5-21 CEPS, ECDPS, AND THE NASDAQ ...................................................................................................... 199 FIGURE 5-22 TECHNOLOGY CPI 1993-2019 ONE-YEAR DIFFERENCED ...................................................................... 202 FIGURE 5-23 TOTAL CONSTRUCTION SPENDING AND TOTAL CONSTRUCTION EMPLOYEES, ONE-YEAR DIFFERENCED .......................................................................................................................................................................... 204 FIGURE 5-24 ALL AVERAGE HOURLY WAGES AND NRCS ......................................................................................... 205 FIGURE 5-25 POLITICAL CONTROL OF THE HOUSE AND SENATE, AND MAJOR INFRASTRUCTURE SPENDING BILLS IN 2005, 2009, AND 2015 ....................................................................................................................................... 208 FIGURE 5-26 PRESIDENTIAL CONTROL AND PCS........................................................................................................ 209 FIGURE 5-27 CEPS, GMPS, ECDPS, AND PRESIDENTIAL CONTROL........................................................................... 210 FIGURE 5-28 CEPS, MINING EQUIPMENT PPI, AND PETROLEUM EQUIPMENT PPI ONE-YEAR DIFFERENCED ............ 212. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 8. DOI:10.6814/NCCU202000410.

(8) FIGURE 5-29 BOX PLOT FOR PCS AND THE HOUSE OF REPRESENTATIVES ................................................................. 214 FIGURE 5-30 BOX PLOT FOR PCS AND THE SENATE ................................................................................................... 215 FIGURE 5-31 SPILLOVERS FROM PCS TO NRCS TO GMPS TO CEPS .......................................................................... 218. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 9. DOI:10.6814/NCCU202000410.

(9) LIST OF TABLES TABLE 3-1 ONE-YEAR DIFFERENCED VARIABLE CODES AND DESCRIPTIONS ...............................................................46 TABLE 3-2 DESCRIPTIVE STATISTICS OF ORDINARY LEAST SQUARE REGRESSION VARIABLES....................................47 TABLE 3-3 CORRELATION COEFFICIENT MATRIX OF ONE-YEAR DIFFERENCED OLS AND VAR REGRESSION VARIABLES ..........................................................................................................................................................49 TABLE 3-4 CEPS, THE NASDAQ COMPOSITE, AND DATA PROCESSING CPI DICKEY FULLER UNIT ROOT TESTS...................................57 TABLE 3-5 DICKEY FULLER UNIT ROOT TESTS FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS..............................................................................................................................58 TABLE 3-6 DICKEY FULLER UNIT ROOT TESTS FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT .........................................................................................................................................................59 TABLE 3-7 DICKEY FULLER UNIT ROOT TEST FOR ONE-MONTH DIFFERENCED NASDAQ COMPOSITE, ONE-MONTH DIFFERENCED IRON AND STEEL PRICES, AND ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES ...60 TABLE 3-8 DICKEY FULLER UNIT ROOT TEST FOR ONE-YEAR DIFFERENCED ECDPS .................................................61 TABLE 3-9 DICKEY FULLER UNIT ROOT TESTS FOR THE SENATE AND HOUSE OF REPRESENTATIVES ONE-YEAR DIFFERENCED ......................................................................................................................................................62 TABLE 4-1 ORDINARY LEAST SQUARE REGRESSION MODELS OF CEPS ONE-YEAR DIFFERENCE ................................70 TABLE 4-2 ORDINARY LEAST SQUARE REGRESSION MODELS OF GMPS ONE-YEAR DIFFERENCED ............................71 TABLE 4-3 ORDINARY LEAST SQUARE REGRESSION MODELS OF ECDPS ONE-YEAR DIFFERENCE .............................72 TABLE 4-4 SELECTION-ORDER CRITERIA TEST FOR ONE-YEAR DIFFERENCED NASDAQ, CONSTRUCTION EQUIPMENT, DATA PROCESSING, AND TECHNOLOGY CPI ...................................................................................74 TABLE 4-5 VAR EQUATION RESULTS FOR THE NASDAQ, CONSTRUCTION EQUIPMENT, DATA PROCESSING, AND TECHNOLOGY CPI ...............................................................................................................................................75 TABLE 4-6 GRANGER CAUSALITY TESTS FOR THE NASDAQ, CONSTRUCTION EQUIPMENT, DATA PROCESSING, AND TECHNOLOGY CPI ...............................................................................................................................................80 TABLE 4-7 LAGRANGE-MULTIPLIER TEST FOR CEPS VAR MODEL 1 ..........................................................................82 TABLE 4-8 LAG ORDER SELECTION CRITERIA TESTS FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS....................................................................................................................85 TABLE 4-9 VAR EQUATION RESULTS FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS......................................................................................................................................... 85 TABLE 4-10 LAGRANGE-MULTIPLIER TEST FOR CEPS MODEL 2 .................................................................................90 TABLE 4-11 VAR GRANGER CAUSALITY RESULTS FOR THE DXY, DIESEL PRICES, AND THE RATIO OF INVENTORY OVER SALES FOR MANUFACTURERS....................................................................................................................91 TABLE 4-12 LAG SELECTION ORDER CRITERIA STATISTICS FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT VAR ........................................................................................................................... 94 TABLE 4-13 VAR EQUATION RESULTS FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT ............................................................................................................................................................................95 TABLE 4-14 LAGRANGE-MULTIPLIER TEST FOR CEPS VAR MODEL 3 ........................................................................99 TABLE 4-15 GRANGER CAUSALITY TEST RESULTS FOR ONE-MONTH DIFFERENCED GDP, ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUNDS RATES, AND ONE-MONTH DIFFERENCED INDUSTRIAL PRODUCTION OF BUSINESS EQUIPMENT .......................................................................................................................................................100 TABLE 4-16 LAG ORDER SELECTION CRITERIA TEST RESULTS FOR ONE-MONTH DIFFERENCED NASDAQ COMPOSITE, ONE-MONTH DIFFERENCED IRON AND STEEL PRICES, AND ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES ......................................................................................................................................105 TABLE 4-17 VAR EQUATION RESULTS FOR ONE-MONTH DIFFERENCED NASDAQ COMPOSITE, ONE-MONTH DIFFERENCED IRON AND STEEL PRICES, AND ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES .106. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 10. DOI:10.6814/NCCU202000410.

(10) TABLE 4-18 GRANGER CAUSALITY TEST RESULTS FOR ONE-MONTH DIFFERENCED NASDAQ COMPOSITE, ONEMONTH DIFFERENCED IRON AND STEEL PRICES, AND ONE-MONTH DIFFERENCED EFFECTIVE FEDERAL FUND RATES ................................................................................................................................................................110 TABLE 4-19 LAGRANGE-MULTIPLIER TEST FOR CEPS VAR MODEL 4 ......................................................................112 TABLE 4-20 LAG ORDER SELECTION CRITERIA TESTS FOR ECDPS, THE NASDAQ, DATA PROCESSING CPI, AND TECHNOLOGY HARDWARE CPI .........................................................................................................................114 TABLE 4-21 VAR EQUATION RESULTS FOR ECDPS, THE NASDAQ COMPOSITE, DATA PROCESSING CPI, AND TECHNOLOGY HARDWARE CPI .........................................................................................................................115 TABLE 4-22 GRANGER CAUSALITY TEST RESULTS FOR ECDPS, THE NASDAQ COMPOSITE, DATA PROCESSING CPI, AND TECHNOLOGY HARDWARE CPI ..................................................................................................................120 TABLE 4-23 LAGRANGE-MULTIPLIER TEST FOR ECDPS VAR MODEL 1 ...................................................................121 TABLE 4-24 LAG ORDER SELECTION CRITERIA TEST RESULTS FOR ECDPS, ONE-YEAR DIFFERENCED DXY, ONEMONTH DIFFERENCED INTEREST RATES, AND ONE-MONTH DIFFERENCED GDP ..............................................124 TABLE 4-25 VAR EQUATION RESULTS FOR ECDPS, ONE-YEAR DIFFERENCED DXY, ONE-MONTH DIFFERENCED INTEREST RATES, AND ONE-MONTH DIFFERENCED GDP ..................................................................................125 TABLE 4-26 LAGRANGE-MULTIPLIER TEST FOR ECDPS VAR MODEL 2 ...................................................................130 TABLE 4-27 GRANGER CAUSALITY TEST RESULTS FOR ECDPS, ONE-YEAR DIFFERENCED DXY, ONE-MONTH DIFFERENCED INTEREST RATES, AND ONE-MONTH DIFFERENCED GDP ...........................................................131 TABLE 4-28 LAG ORDER SELECTION CRITERIA FOR ECDPS, IRON AND STEEL PRICES ONE-MONTH DIFFERENCED, INDUSTRIAL PRODUCTION ONE-MONTH DIFFERENCED, AND DIESEL PRICES ONE-YEAR DIFFERENCED...........134 TABLE 4-29 VAR EQUATION TEST RESULTS FOR ECDPS, IRON AND STEEL PRICES ONE-MONTH DIFFERENCED, INDUSTRIAL PRODUCTION ONE-MONTH DIFFERENCED, AND DIESEL PRICES ONE-YEAR DIFFERENCED...........135 TABLE 4-30 LAGRANGE-MULTIPLIER TEST FOR ECDPS MODEL 3 ............................................................................139 TABLE 4-31 GRANGER CAUSALITY TEST RESULTS FOR ECDPS, IRON AND STEEL PRICES ONE-MONTH DIFFERENCED, INDUSTRIAL PRODUCTION ONE-MONTH DIFFERENCED, AND DIESEL PRICES ONE-YEAR DIFFERENCED...........140 TABLE 4-32 OLS REGRESSION IMPACTS OF POLITICAL CONTROL .............................................................................145 TABLE 4-33 LAG ORDER SELECTION CRITERIA TESTS FOR THE HOUSE, SENATE, AND CEPS ....................................146 TABLE 4-34 VAR EQUATION RESULTS FOR THE HOUSE, SENATE, AND CEPS ...........................................................147 TABLE 4-35 CORRELATION COEFFICIENT MATRIX OF MACROECONOMIC INDICATORS AND POLITICAL CONTROL DUMMY VARIABLES ..........................................................................................................................................150 TABLE 4-36 POLITICAL CONTROL, CEPS, NRCS, AND PCS VAR EQUATION RESULTS .............................................152. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 11. DOI:10.6814/NCCU202000410.

(11) Appendix C: List of Terms AIC—Akaike’s information criterion AICs—Advanced industrial countries CEPs—Construction equipment prices CPI—Consumer price index DXY—Trade weighted dollar index ECDPs—Excavators, cranes, and draglines prices EPA—Environmental Protection Agency FPE—Final prediction error. 政治大 GDPPC—Gross domestic product per capita 立 GDP—Gross domestic product. ‧ 國. 學. GFC—Global Financial Crisis GNP—Gross national product GMPs—General machinery prices. ‧. HQIC—Hannan and Quinn information criterion. Nat. sit. y. IRF—Impulse response function. io. al. n. LR—Likelihood ratio. er. LDCs—Lower developed countries. Ch. NICs—Newer industrialized countries. engchi. NRCS—Non-residential construction spending. i n U. v. OLS—Ordinary least squares ORIF—Orthogonalized impulse response function PCS—Public construction spending PPI—Producer price index R&D—Research and development SBIC—Schwarz Bayesian information criterion VAR—Vector autoregression VARX—Vector autoregression with exogenous variables 12. DOI:10.6814/NCCU202000410.

(12) Chapter 1 1.. Introduction Equipment prices are an important component to the construction industry which is a. reflection on the overall macroeconomic condition of an economy. Because of this and other factors, it can be surmised that equipment pries can be used as a key indicator of economic activity (Gerrard, 1978, pg. 241). Additionally, the price of the equipment may determine whether or not the piece of industrial machinery should be purchased. Even though the decision to purchase a new. 治政大 by the broader macroeconomic macroeconomic conditions since small businesses can be impacted 立 environment or macroeconomic policies such as interest rates and cyclical components such as piece of equipment is a microeconomic issue, the price of the equipment is subject to. ‧ 國. 學. recession (Ibid). Moreover, construction equipment prices (CEPs) are, from a theoretical point of. ‧. view, a reflection of many complex components of industrial production, employment, spending, technological advancements, and growth at the equilibrium point in which buyers are willing to. y. Nat. er. io. sit. purchase expensive new capital (Shea, 1993, pg. 5). If prices are too high, then investment into construction projects may be stifled. Because of this, it is not then surprising that equipment prices. n. al. Ch. i n U. v. are negatively correlated with investment in developed countries (Ibid., pg. 26). Since this is the. engchi. case, equipment prices can elucidate important findings on what future investment levels may be. How then can one determine prices of construction and related equipment? Neoclassical macroeconomic theories suggest that the price of capital is the equilibrium point of the downward sloping convex line of capital demand and the vertical line of capital supply (Goolsbee, 1998, pg. 140). In spite of this, capital demand capital supply are not the only components which influence the final production price of industrial machinery and equipment. Of course, neo-classical models also take into consideration the price of inputs such as interest rates (Gerrard, 1978; Fama and Gibbons, 1982). 13. DOI:10.6814/NCCU202000410.

(13) Also, a crucial component of capital price production is oftentimes ignored which is technological advancement. During the entirety of the 20th century, theoretical models could afford to do so since the pace of technological advancement was relatively slow compared to what it is today. But unfortunately, this is no longer the case. The pace of digitalization, the implementation of algorithmic analysis, or the use of robotics, automation, artificial intelligence, and the overall computerization of industries has spilled over into machinery markets. This is the case in regard to technological implementation in the production of machinery and within the machines. 治政大 production in conjunction with price of machinery as can be seen by the increase in manufacturing 立 themselves. Without a doubt, technological advancement is continuing to influence the production. the large decrease in manufacturing jobs. Is this then also true for construction equipment? What. ‧ 國. 學. then is the impact of technology on CEPs? Moreover, what then are the correlates of CEPs in. ‧. general?. In spite of this key development in equipment markets, few academic studies investigate. y. Nat. er. io. sit. the impact of technological development on the cost of equipment production. Most of the literature which does address the correlates of equipment prices focuses on agricultural machinery. n. al. Ch. i n U. v. (Rada and Valdes, 2012; Kumar and Jashi, 2014; Zou and Qiu, 2017). This is likely the case given. engchi. that agricultural equipment was the first type of machinery to be automated (Ibid). On the contrary, construction equipment has not suffered from automation (in terms of the labor market and the operationalization of construction projects) and requires a large degree of human interaction. Three main sets of regressions were conducted with differing dependent variables. The first set of regressions conducted used one-year differenced data measuring CEPs against a host of variables including technological factors, macroeconomic factors, and commodity inputs. The second set of regressions used all of the same models as the first set of regressions in addition to. 14. DOI:10.6814/NCCU202000410.

(14) one-year differenced data, but with a different dependent variable of general equipment prices which includes within its measurement construction-related equipment. While not entirely the same, the two variables reflect different aspects of construction where general equipment reflects the prices of construction-related equipment prices. The third set of regressions uses excavators, cranes, and dragline prices (ECDPs). The findings of this paper can be put into four main boxes. First, technology has an in general negative net impact on CEPs in the long-run. Second, supply of construction equipment. 治政大(NRCS) and speculative demand of demand as measured by non-residential construction spending 立 and general equipment is much less impactful on the price of equipment as opposed to the variable. as measured by the ratio of inventory over sales of the manufacturing industry. Third, these. ‧ 國. 學. relationships are much stronger after the 2000-2001 Tech Bubble as opposed to before it. This is. ‧. demonstrated statistically through the implementation of variables which only use data from after the Tech Bubble where coefficients are dramatically increased across variable typologies. Fourth,. y. Nat. Representatives tends to increase CEPs and PCS.. n. al. Ch. er. io. sit. political findings were made as well, primarily that Republican control of the House of. i n U. v. A number of VAR models were also used in the method section of this thesis. The findings. engchi. of these VARs were mostly consistent with the OLS findings. A few exceptions to this general finding include the difficulty in using effective federal fund rates since one-month differencing was required in order to fit a valid VAR with this variable. Moreover, data processing CPI did not entirely provide useful results in the VAR models as a result of the 90 percent confidence intervals being both above and below the zero-point reference line. With this study, many correlative relationships were uncovered between different equipment price indices and a number of financial and macroeconomic indicators, primarily that. 15. DOI:10.6814/NCCU202000410.

(15) investment, and technology lower prices while commodities, industrial production, and interest rates elevate them. While the overall relationship between macroeconomic variables and price remains complicated, the extensive methods used throughout the duration of this thesis provide a clearer framework through which to analyze CEPs, GMPs, and ECDPs which are mostly ignored throughout the academic literature. The impact of technology is expounded upon in depth. The degree to which supply impacts price demonstrates a troubling relationship on the nature of the construction industry writ large. A better understanding is created on the nature of policy-making. 治政大this study fills a great void in the as it pertains to spillover effects from NRCS to PCS. Overall, 立 as well as is evident on the section covering political impacts of control over the House and Senate. literature examining equipment price correlates and provides an ample amount of room to further. ‧ 國. 學. expound upon these ideations in similar methodologies.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 16. DOI:10.6814/NCCU202000410.

(16) Chapter 2 2. Literature Review When prices increase – commonly referred to as inflation – it is generally a sign for economists or analysts perceived to be a positive indication for the health of a functioning economy when restrained. The latter point on restraint is an important caveat. Most central banks of developed economies, such as the United States’ Federal Reserve or the European Central Bank, tend to target inflation at two percent (Almeida and Goodhart, 1998). In fact, low inflation targets. 治政 exception of some targeting employment such as the Reserve 大 Bank of New Zealand (Turner, 2017, 立 pg.3) and overall financial stability such as the Bank of England (Dikau and Volz, 2019). Since are one of the few mandates most central banks around the world are governed under with the. ‧ 國. 學. most inflation indicators are cointegrated, focusing on construction price inflation is justified in. ‧. this regard. lao. Another justification for the focus on construction equipment and equipment prices in. y. Nat. er. io. sit. general resides in the association between the construction industry and economic growth. To that end, some research has been devoted to studying the relationship between construction and. n. al. Ch. i n U. v. economic growth. Specifically, construction capital expansion was found to undergo a correlation. engchi. with nominal and per capita GDP growth than any other sector of the economy (Giang and Feng, 2011). Moreover, the academic literature takes note that there is also a significant difference between the relationship of economic growth generally speaking and the construction industry generally regarding developing economies (Lopes et al., 2002) and developed economies (Bon, 1992). Particularly, the relationship between construction as a share of gross national production (GNP) and GNP per capita is non-linear. For the most part, underdeveloped economies have little construction, semi-developed economies enjoy much construction, and developed economies have much less in what is known as the Bon Curve (Ibid.). Although newly industrialized countries 17. DOI:10.6814/NCCU202000410.

(17) (NICs) enjoy a larger share of construction in GNP as compared to advanced industrial countries (AICs), the share of construction in GNP of AICs is larger than that of less developed countries (LDCs) (see Figure 2-1). Moreover, the empirical evidence suggests that construction as a share of GDP still remains at around 6 percent for the United States (Wang, 2019).. 立. 政治大. ‧. ‧ 國. 學 er. io. sit. y. Nat. n. al. i n C Figure 2-1 The Bon Curve. FromhBon e n(1992). gchi U. v. However, the relationship between GDP growth and the construction growth is more complicated than this generally theoretical suggestion with closer examination. Specifically, the directionality of the causal nature between economic growth and construction industry is for the most part inconclusive. For instance, research using data from Hong Kong found that GDP growth caused construction industry growth (Tse and Ganesan 1997) while other research using data from Western Europe found that there was weak evidence pointing to the impact of GDP growth on infrastructural investment (Wilhelmsson, and Wigren, 2009). However, Jones (1994) found that 18. DOI:10.6814/NCCU202000410.

(18) there was a negative relationship between GDP growth and relative machinery prices which appeared to be true for developed and developing countries from 1960 until 1985. Jones also found that there was an exponential convex relationship between GDP per capita (GDPPC) and relative machinery prices in developed and developing economies during the same time period (pg. 363). In spite of these general findings regarding the construction industry, only a few empirical studies specifically examine variables which correlate with CEPs, ECDPs, or GMPs. The reason behind this is likely due to the fact that it is assumed through the neo-classical model of investment. 治政大 some studies examining the of capital demand and the vertical line of capital supply. Indeed, 立. that capital prices are implied through the equilibrium of the relationship between the convex line. correlates of capital prices even go so far as to assume that total equipment sales are effectively. ‧ 國. 學. the same as or a proxy of capital demand (Toole, 1998; Ruddock et al., 2010).. ‧. Nonetheless, the Cobb-Douglas production function and other production functions related to it are also commonly used to set a parameter of the variables which correlate with the price of. y. Nat. er. io. sit. capital (Douglas, 1978). These variables include items such as the stock of capital, the unemployment rate, and others. Critically though, nearly every popularly used production function. n. al. Ch. i n U. v. ignores technological growth and a function for the nature of growth in efficiency that results from. engchi. technological innovation. When using this function and others like it, other problems arise as well when applying to empirical models econometrically. This includes the fact that capital stock figures are only measured at a yearly rate forcing very long-term time series studies to be used. Doing so tends to ignore how different decades are accentuated by multiple unique economic characteristics and how correlation coefficients can dramatically change over time through the different business cycles. This problem is only exacerbated by the increase in the rapidity of technological innovation in the past few decades. Second, it is difficult to determine which. 19. DOI:10.6814/NCCU202000410.

(19) barometer of technology to use since such an indicator is difficult to quantify. For instance, the number of pieces of equipment which are computerized on the marketplace is not publicly recorded information. Moreover, the few empirical studies that look at the correlates of machinery prices fail to even use technological variables as explanatory factors of equipment prices. Clearly though, the most glaring modern problem with the traditional approaches of determining CEPs, ECDPs, or GMPs is the relationship between increasing technology and the price of capital to the point in which some have argued that the popularized production functions. 治政 which do take into consideration technological progress, yet大 these as well do not satisfy critical 立 should be thrown out completely (Labini, 1994). Other production functions have been proposed. empirical themes. An alternative model suggests that technological progress can be endogenous to. ‧ 國. 學. the model while augmenting certain aspects of the traditional production functions. Technological. ‧. progress cannot be held constant. This is evident in the previous models. Moreover though, it must be taken into consideration how technological progress impacts the primary inputs of production. y. Nat. er. io. sit. functions, labor and capital, to produce growth. For instance, neutral technological growth models using Solow-neutral technological growth model prove that technological growth can be. n. al. Ch. i n U. v. implemented into neo-classical production functions, but only if that technological growth keeps. engchi. intact the relative labor to capital ratios. In other words, technological growth would not reduce or increase labor relative to capital, a trend which is clearly not consistent with the empirical ledger on labor’s impact from technology. However, the Harrod-neutral growth model is unconcordant with aspects of the Solow-neutral growth model, particularly that that the Solow-neutral technological growth only augments capital while the Harrod-neural technological growth model only augments labor. However, the empirical record suggests that technological growth is not entirely only labor augmenting or capital augmenting. In fact, it is probably both.. 20. DOI:10.6814/NCCU202000410.

(20) While it may appear to be a simple theoretical foundation that efficiency in producing machines would decrease the price, the complexities of empirical macroeconomic conditions in relation to CEPs, ECDPs, or GMPs is difficult to pin down to a simple few number of variables. Because of this, more complex models are likely required in order to more accurately understand their fluctuations. Interest rates for example are traditionally viewed as the cost of capital. This is still true, but only to a certain extent. Clearly, GDP impacts many other economic indicators not underneath the umbrella of its core components. Traditionally, it is also viewed that input costs. 治政 this is the case is up for debate as is the directionality of the 大 relationship between these variables 立. such as labor and commodities likely play a role in prices as well. However, the degree to which. and the aforementioned variables of technology and GDP. Much more basic variables could play. ‧ 國. 學. a role such as supply and demand, but it may be the case that investment is the most important. ‧. factor of all. This literature review will then go over these different variables to assert what the academic literature and what theoretical foundations have to say about the relationship between. y. Nat. er. io. sit. capital prices and their correlates. Moreover, it will take into consideration what traditional theory has to say on the matter was well.. n. al. 2.1. Interest Rates and InvestmentC h. engchi. i n U. v. As mentioned, neo-classical macroeconomic theory generally asserts that the price of any good or service is devised from the implied equilibrium of supply and demand of such goods or services. Specifically, the literature suggests that capital goods can be defined as, “the physical objects (factories, machines, apple trees) produced by inputs now and used to produce outputs in the future,” (Friedman, 2019, pg. 343). Fundamentally, the purchasing of a piece of machinery is an investment, but the price of the investment made is based on multiple variables. Primarily, this study is focused on the price of production and is therefore not concerned with a large market of 21. DOI:10.6814/NCCU202000410.

(21) equipment and machinery which is arbitraged. Nonetheless, traditional neo-Keynesian theoretical assumptions suppose that a firm’s decision to invest in a piece of machinery is predicated upon interest rates which, when lower, incentivize firms to investment and when higher to hold off on borrowing to invest. The reason why interest rates are so important theoretically as price determinants of capital is that they are fundamentally, at a theoretical level, assumed to be equal to the price of capital (Ibid., pg. 344). This is the case because monetary capital is either invested in physical capital (factories or machines) or savings. Therefore, in the long-run the cost of capital. 政治大. must be the interest rate either earning interest on savings or suffering from interest on borrowing and subsequent investing.. 立. The empirical literature also suggests that interest rates are quite important in the. ‧ 國. 學. movement of equipment prices. Monetary policy can be aimed at addressing the distortions of. ‧. factor prices such as wages of manufacturing employees through interest rate adjustment (Giang and Feng, 2011, pg. 120). Additionally, interest rates were found to be statistically significant with. y. Nat. er. io. sit. capital investment (Fama and Gibbons, 1982). Moreover, a strong school of thought on monetary theory asserts that monetary policy can cause overall goods inflation if money supply expansion. n. al. Ch. i n U. v. is larger than gross domestic product (GDP) (Bernanke and Blinder, 1988). Because of this, it. engchi. stands to reason that monetary policy could also impact CEPs given the impact that the theoretical foundation suggests interest rates have. However, little research has focused on the impact of monetary policy specifically on the price of CEPs. Since this is the case, it is difficult to assert one way or another the degree to which monetary policy empirically impacts CEPs, but the literature generally asserts that there is a positive relationship between interest rates and capital prices (Goolsbee, 1998). On top of this, the primary theoretical foundation additionally asserts that interest rates may have a positive impact on CEPs since lower interest rates translate into lower. 22. DOI:10.6814/NCCU202000410.

(22) borrowing costs which in turn translates into more investments in capital (Fama and Gibbons, 1982). In general, analysts assert that relative equipment and machinery prices are highly correlated with investment and economic growth (Restuccia and Urrutia 2001; Jones 1994; Sarel 1995; Lee 1995; DeLong and Summers 1991). Indeed, the empirical record of cross-country analyses asserts that capital prices are more expensive in underdeveloped economies, less expensive in developing economies, and even more less expensive in developed economies (Aslan. 治政大Moreover, it has been argued that trade openness and the elasticity of labor productivity (Ibid.). 立 et al., 2019). Research suggests that around 50 percent of this relationship can be accounted by. the decreasing of capital prices has spurred investment and not the other way around (Ibid.).. ‧ 國. 學. However, the directionality of these relationships are under dispute as others have suggested that. ‧. investment in capital markets impacts prices and not the other way around (Jones, 1994; Wilhelmsson, and Wigren, 2009).. y. Nat. er. io. sit. Models have suggested that higher levels of investment are correlated with higher levels of equipment prices (Kogan and Papanikolaou, 2014, pg. 711). On theoretical grounds though, this. n. al. Ch. i n U. v. notion appears to be inconsistent with neo-Classical models which suggest that technological. engchi. development would reduce equipment production prices since technological development should increase efficiency in production. Capital spending in the 1990s was in part due to due plummeting computer prices and a surge in technology equities (McCarthy, 2001, pg. 1). Research has also found that from 1870 to the 1990s, productivity growth has been positively correlated with investment in equipment (De Long, 1995). Investment in equipment has also been posited to be positively correlated with economic growth and that industrialization is driven by human capital accumulation (Temple and Voth, 1998). Moreover, investment in capital goods is negatively. 23. DOI:10.6814/NCCU202000410.

(23) correlated with capital prices as has been demonstrated by recent research on the topic (Collins and Williamson, 2001). Since this is the case, this thesis posits that the relationship between investment in technology (proxied by the NASDAQ Composite) and CEPs is negative. 2.2. Growth, Supply, Demand, Labor, and Commodities Studies on price changes are more so focused on specific indicators such as CPI, PMI, or industrial production as opposed to machinery prices. This is likely due to the fact tht the construction industry is much less focused on in macroeconomic studies since the industry tends. 政治大 relationship between them and GDP 立 figures. Additionally, the literature on machinery prices tends. to lag the business cycle. This is also true for CEPs which likely accounts for the negative. ‧ 國. 學. to investigate the price of used machinery and the degree to which machinery prices of already purchased machinery is valued vis-à-vis depreciation (Hulten and Wykoff, 1980; Day and. ‧. Benjamin, 1991; Lanteri, 2018). While the price of used equipment is an important subject. sit. y. Nat. deserving attention, this topic is already covered at great length in the aforementioned literature.. er. io. Therefore, examining the correlates of used equipment prices and rental rates is outside the. al. n. iv n C h e nCEPs Research on the relationship between h itheUdemand for construction equipment g cand. parameters of this thesis.. has found that domestic demand of heavy equipment as measured by the volume of heavy equipment sales and domestic production of heavy equipment is cointegrated in a positive manner (Sidharth et al., 2015, pg. 7). In other words, it may be the case that production or the supply of construction equipment plays a role its price. This is one of the key components of the neoClassical models; where supply meets demand is where the equilibrium price can be found allegedly. However, this is not the case in monopolistic or oligopolistic competition where firms. 24. DOI:10.6814/NCCU202000410.

(24) have more power on setting price regardless of supply. In spite of this, if demand is not present, then it is more difficult to support non-competitive pricing. Economic growth has also been suggested to be correlated with equipment prices. Other research examined how relative capital goods prices are strongly and negatively correlated with investment rates and GDPPC at a statistically significant rate (Collins and Williamson, 2001). CEPs have been found to be closely related to real GDP (Lanteri, 2018, pg. 70). Other research on the relationship between GDP growth and the construction industry has been previously noted.. 治政大is the case historically. However, is an important reflection of the overall economy. At least, this 立 Overall, it is clear to see that the relationship between overall economic growth and durable goods. since the US economy is much less based upon manufacturing for economic growth, it is less so. ‧ 國. 學. the case today than what it was decades ago when the majority of research using growth levels and. ‧. machinery prices as variables.. There may also an important relationship between labor and equipment prices. Since. y. Nat. er. io. sit. manufacturing of complex construction equipment is not yet fully automated, labor costs remain one of the highest costs for producers of durable capital goods. Perhaps because of this, Hu and. n. al. Ch. i n U. v. He (2014) assert that used CEPs are simply a function of labor costs and material costs relative to. engchi. the quality of the construction equipment in terms of depreciation, durability, and efficiency (pg. 4). However, it should also be taken into consideration that these costs within their specific production function were not only for new equipment, but used construction equipment and that secondhand markup costs from arbitration are implemented. Moreover, it also needs to be recognized that used construction equipment cost functions are entirely separate from those of newly produced construction equipment costs, the former of which does not take into consideration labor costs. Overall, the impact of labor costs on the production of new machinery may be small.. 25. DOI:10.6814/NCCU202000410.

(25) 2.3. Oligopolistic Competition in the US Construction Industry Construction equipment production market in the US is effectively an oligopoly (Domberger and Fiebig, 1993; Milford, 2015). In fact, Douglas (1976) predicted that prices would be higher in spite of higher production in monopolistic or monopolistic-like industries (pg. 909). Indeed, the US company Caterpillar dominates by itself 40 percent of the entire US construction equipment market share (Saidani et al., 2019). Along with a few other companies such as Komatsu, Deere, and Volvo, only 30 percent of the market is owned by small businesses (Ibid.). Moreover,. 政治大 were attempting to monopolize立 the US construction equipment market by not allowing Chinese Caterpillar, Komatsu, and Volvo were all sued by a Chinese competitor who claimed that they. ‧ 國. 學. machines to be purchased directly from manufacturers on the online construction equipment auction website Iron Planet (Milford, 2015). While an argument can be made over whether or not. ‧. the US construction equipment industry is a monopoly or an oligopoly, there is little doubt that. sit. y. Nat. very few companies control the market and therefore have more power when it comes to setting. er. io. the price of their equipment.. al. iv n C by decreasing the quantity of products h made e nandg therefore c h i Uincreasing prices on these products. n. As can be seen by Figure 2-2, oligopolies gain what is referred to as ‘supernormal profit’. This also incurs a deadweight loss which is essentially a cost incurred upon a society or consumers from an inefficient method of production. It is because of the aforementioned information on market share and online availability which provides a bit of justification on the claim and lawsuit that Caterpillar, Komatsu, and Volvo were engaging in oligopolistic actions. Clearly, consumers of construction equipment would be able to purchase machinery much more cheaply if they were not forced to go through middle-men dealers who purchase from the producers and then sell at an arbitraged markup to the contractors who use the machinery themselves. Prices increase and 26. DOI:10.6814/NCCU202000410.

(26) deadweight loss is incurred. Unfortunately for the consumers of construction equipment, the lawsuit was flawed in spite of the evidence that Chinese machines were unable to be sold on the producer-to-consumer online auction website Iron Planet.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 2-2 Oligopolistic Price Manipulation and Profit Taking. However, there is a problem with this supposition which is primarily that monopolies and oligopolies for that matter both generate their profit from reducing the quantity of products as opposed to simply raising prices. Oligopolistic profit occurs in a region between average cost, demand, marginal cost, and two points on the y-axis of price. The quantity of production occurs where marginal cost meets marginal revenue. Because of this, the only rational reason behind why 27. DOI:10.6814/NCCU202000410.

(27) there is a positive relationship between equipment prices and supply (and subsequently the positive relationship between industrial production with the ratio of inventory over sales) must be that producers are merely replying to an increase in demand by increasing supply. Moreover, the increase in supply is slightly smaller than what is demand so that prices increase, but not at true equilibrium. If the increase in supply were to match demand, then the price would stay the same. Moreover, the order of this relationship is a necessity as far as explaining this phenomenon primarily due to the fact that if there were first an increase in supply in anticipation of an increase. 治政大increase in demand. have been reduced had the increase in supply been equal to the 立. in demand, the prices would first decrease. Therefore, prices increase from P0 to P1, but could. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 2-3 How Oligopolies Can Increase Price While Also Increasing Supply. 28. DOI:10.6814/NCCU202000410.

(28) However, just because the US equipment industry is not a monopoly, it could be the case that this targeted increase in supply which does not meet true equilibrium is collusion (technically still a limit on supply, but just in the change of supply). Therefore, it could still be argued that taking time and the change in time, there is still a decrease in the supply since firms are failing to increase supply sufficiently enough to meet demand. Since this is the case, it must still be considered that the US equipment industry is still suffering from oligopolistic conditions. 2.4. Equipment Prices and Other Financial Market Indices. 治政 It could be the case that machinery prices are also heavily 大 impacted by commodity input 立 costs such the cost of fuel (Cartwright et al., 1989, pg. 79). Most manufacturing plants require ‧ 國. 學. fossil fuel-based energy in order to power these plants, even in the advanced economy of the. ‧. United States (Ram et al., 2017). Additionally, Equipment prices in the power sector were using data from January 1996 until December 2003 and data from January 2004 through December 2007. y. Nat. io. sit. found that price increases were primarily due to the increase in the demand for oil (Pauschert,. n. al. er. 2009, pg. 3). This increase in oil peaked during the bubble of global asset prices commonly referred. Ch. i n U. v. to as the 2008 GFC. These financial trends also happened to coincide with an overall rise in US. engchi. CEPs (Federal Reserve Bank of St. Louis, 2019) which suggests that commodity prices are much more reactionary in the short-term to financial conditions than equipment prices. However, research has found that there is only a two percent increase in prices of construction machinery from oil shocks which, compared to other asset shocks, was substantively insignificant (Lee and Ni, 2002). Since the results of the data provides different results over different periods of time, it is difficult to determine the degree to which petroleum products or the trade weighted dollar (DXY) – impacts CEPs.. 29. DOI:10.6814/NCCU202000410.

(29) Other inputs such as steel prices also likely have a very strong impact on CEPs. Heavy equipment and other machinery are primarily made out of steel and iron which have in the past decade suffered from dramatic volatility in spot prices (Dwyer et al., 2011, pg. 49). Furthermore, commodity prices such as steel have also been found to be related to machinery prices in the power industry (Pauschert, 2009). This may therefore also be the case as well in the construction equipment market. Without a doubt though, the price is steel is strongly significant in the overall cost of construction (Masur et al., 2016). Steel is undoubtedly a major input in the building of. 治政大spillover effects. that the price of steel is correlated with CEPs from construction 立. roads, bridges, buildings, and other structures. Because of this, it would not be surprising to find. There may also be an impact on CEPs related to the exchange rate. While the production. ‧ 國. 學. of construction equipment itself is not necessarily subject to exchange rates, the price of inputs. ‧. which are imported into the US and implemented or used in the production of construction equipment are of course impacted by the exchange rate. Goods and the cost of inputs on the. y. Nat. er. io. sit. international market are subject to foreign exchange costs. When the US dollar is strong relative to its trading partners currencies, then the cost of imports is lower and conversely these costs are. n. al. Ch. i n U. v. higher when the dollar is weaker. As a general rule of thumb though, a strong DXY is a reflection. engchi. that investors or potential investors (regardless of what industry they are in) prefer to hold onto those dollars as opposed to spending them. Furthermore, there is a long-standing inverse relationship between the DXY and commodity prices (Bai and Koong, 2018). Moreover, when the DXY is stronger, then international goods may be more competitive to import against domestically produced goods. Undoubtedly, US imports have incurred the cost of foreign exchange for several decades and the relationship between nominal imports and the exchange rate is strong and statistically significant (Hooper and Mann, 1989). Additionally, it has been found that before the. 30. DOI:10.6814/NCCU202000410.

(30) 2008 GFC, machinery prices in the US were correlated with the DXY (Pauschert, 2009). Considering these implications in the literature, it may be the case that this variable could be a determinant of CEPs. Even though the empirical record is relatively clear on its relationship between capital inflation or deflation and financial markets, the reason as to why this is remains murky. The relationship between capital inflation or deflation and equity markets is clearer as it there are more direct components to the influx of money into firms which allows for them to invest more into. 治政大markets are correlated with one and these prices? It seems to be the case that many financial 立 capital. However, why would there then be strong relationships between other financial markets. another. As previously mentioned, the DXY and diesel or oil are highly negatively correlated.. ‧ 國. 學. Many other correlations exist as well though.. ‧. Probably the most notable correlation in financial markets is that between currencies and equities which could impact equipment prices. Without a doubt, the strength of the currency is a. y. Nat. er. io. sit. reflection of the demand for that currency. However generally speaking, two main competing theories provide a framework through which to analyze the relationship between equity markets. n. al. Ch. i n U. v. and foreign exchange markets. The first is the goods market hypothesis (Dornbusch and Fischer,. engchi. 1980) which asserts that exchange rate fluctuations have a significant impact on the competitiveness of firms which conduct business internationally (most larger firms in today’s world) which in turn impacts their equity prices. If a local currency is weakened, then its goods are more enticing for foreign buyers which leads to higher sales. Naturally then, the value of these exporting firms increases. Exporters also are affected by price fluctuations of exchange rates in terms of their future payables that are denominated in foreign currencies. This is also how the. 31. DOI:10.6814/NCCU202000410.

(31) appreciation or depreciation of the local currency could decrease or increase profits. In brief, currencies lead equities. The second main explanatory ideation on the relationship between foreign exchange and equity markets is relatively contrary to the first ideation. That is to say, those of the ‘portfolio balance approach’ (Branson 1980; Frankel and Dickens, 1983) reminds us that money itself is an asset with an intrinsic value and is used as a portion of a portfolio. Because of this, foreign exchange rates are also determined by market forces. As can be simply understood through the. 治政大is more sought after for the use of determined by capital inflows and outflows. When a currency 立 basic framework of supply and demand, a currency’s nominal rate can fundamentally be. purchasing stocks, bonds, or for conducting business in the country, the value of the local currency. ‧ 國. 學. strengthens. In contrast, when foreign capital flows out of international equity or bond markets. ‧. and convert the local currency back to their foreign currency or into another currency, the value of the local currency is weakened. Foreign investment tends to increase over time through the model. y. Nat. er. io. sit. of portfolio diversification as well which has led to the in general weakening of the dominant global currency, the US dollar since the 1980s. In sum, equities lead currencies.. n. al. Ch. i n U. v. It may also be the case that both equities lead currencies at some points in time and then. engchi. the correlation changes for the currency to lead equities. Mok (1993) uncovered a dual directional relationship between causal relationship between global currencies and equities in Hong Kong, a city with one of the most advanced and liquid financial markets in the world. Moreover, this relationship between equities and currencies is not always constant. Adding onto this initial finding, a causal relationship between stocks and currencies was found to be present in the short run, but not the long run (Bahmani-Oskooee and Sohrabian, 1992; Nieh and Lee, 2001).. 32. DOI:10.6814/NCCU202000410.

(32) Regardless of whether currencies lead equities or whether equities lead currencies, it can be generally surmised that when the demand in a particular currency rises its strength rises which is in turn likely the manifestation of the desire for investors to either invest into that economy whether it be in equities (the most common financial investment), that country’s bonds, or capital in that country in order for business to establish infrastructure. Spillovers of volatility between both markets may increase the international portfolio risk faced by international investors. This reduces the opportunities from international diversification and disturbs the asset allocation. 治政大 both markets. for currencies, leading to some degree of interdependence between 立. decisions. Rapidly increasing international equity investments creates a higher supply and demand. ‧. ‧ 國. 學. 2.5. Machinery or Equipment Prices or Inflation and Technological Innovation. sit. y. Nat. Although the literature on specifically CEPs, ECDPs, or GMPs suffers from a lack of depth,. er. io. the literature on innovation and technology in the construction industry is significantly more. al. n. iv n C h eprices quantitative correlation between equipment i Uand the implementation of modern n g cin hgeneral elucidating. However, in spite of the insight which can be gained from understanding the. computer technology into construction equipment, such a task remains incredibly difficult to accomplish. This is primarily due to the fact that such data—the implementation of computer technology into construction equipment—is simply not open source and readily available to the public. Such is true also for data related to robotics and manufacturing of industrial goods. Moreover, how one would quantify such a variable is also difficult to ascertain. Because of this, it is nearly impossible to point to a specific body of research which specifically investigates this relationship. 33. DOI:10.6814/NCCU202000410.

(33) In spite of this, there are still some studies in the academic literature which investigate the relationship between technological advancements and the construction equipment industry. For instance, Adriti et al. (1996) asserts several findings including that technological development in earth-moving equipment proceeded steadily over several decades from the 1960s to the 1990s. Moreover, this research found that technological advances in construction equipment has not historically or merely been confined to the construction equipment industry. Rather, technological development in the construction industry enjoys a spillover effect from other machinery industries. 治政大 yearly was an indicator of generally found that the introduction of new equipment as measured 立. and innovation tends not to start first in the construction equipment industry. Finally, this research. technological development. However, several problems exist with this conceptualization.. ‧ 國. 學. Primarily, it does not take into consideration imitation or the fact that previous machines models. ‧. may not have significant technological advances as the previous ones. Other research has found that technological innovation in the construction industry is quite slow compared to the. y. Nat. er. io. sit. manufacturing sector (Laryea and Ibem, 2014).. Importantly, Toole (1998) found that increasing technological development in construction. n. al. Ch. i n U. v. equipment led to a decrease in prices of construction equipment. Furthermore, this was also found. engchi. to be the case in not only the resale price of construction equipment that was new and used, but also this was the case for the cost of construction equipment production (Ibid.). Nam and Tatum (1992) also found that technological advancements in construction equipment led to price reductions. However, they contributed that this price was reduction was not incremental like the advancement of technology, but was actually dramatic compared to the advancement of technology (Ibid.) In other words, while technological advancement may be incremental, the price reduction in construction equipment was much stronger than was the pace of development. This is. 34. DOI:10.6814/NCCU202000410.