Does trade with China benefit the U.S.?

Forrest Scott Brinkley

Economics Major

December 7, 2006

A hot button issue in contemporary politics, the United States’ current trade with

China is often blamed for American economic woes. The biggest issue is probably the massive American manufacturing job losses that started in the 1980s and continue through today. Our trade with China is also heavily unbalanced. Yet the question remains as to whether or not the long-run economic impact of trade with China will be good or bad for the U.S. macro economy. This paper examines, through tools such as the regression and correlation, the relationship between three variables related to U.S. foreign trade and its economy.

I.  Introduction

The controversial trade situation between China and the U.S. has been mentioned at least implicitly in the speeches of every major politician since the 1990s and perhaps before then. Union workers are upset about their jobs being seemingly displaced by the increased flow of cheap Chinese-made goods through U.S. ports. Wal-Mart Stores, INC. has been ostracized by many for stocking and selling such goods. Every major news media outlet has, at least once, mentioned the huge trade deficit between the U.S. and China. Are the long-run effects better for America and Americans? For instance, is the trade with China helping to keep the American CPI from rising? How does trade affect corporate profits? As much as some are angered by huge corporate profits, a capitalist economy is dependent upon them. These important economic factors will be analyzed using SAS to see how and if they relate to the U.S.-Chinese trade situation.

II.  Data & Statistical Tools Used

Data used for this project included CPI, corporate profit, and trade data from various U.S. government agencies.[1] The data points were extracted and/or downloaded carefully from the respective agencies’ websites and then compiled and manipulated as necessary in separate Microsoft Excel worksheets to maintain the integrity of the original data. Some modification of format was necessary to make the data transition smoothly into SAS, the statistical analysis package of choice for this project. Additionally, all data had to have a common variable, “year” in this case. Once in SAS, the data points were all merged into a SAS dataset which could be modified and used by SAS for analysis without manipulating the original data files. Table 1 below is the result of that merging.

Year / Annual % Change in CPI / Corporate Profits / Annual Balance in U.S.-Chinese Trade
1985 / 3.6 / 330.3 / -6
1986 / 1.9 / 319.5 / -1664.7
1987 / 3.6 / 368.8 / -2796.3
1988 / 4.1 / 432.6 / -3489.3
1989 / 4.8 / 426.6 / -6234.3
1990 / 5.4 / 437.8 / -10431
1991 / 4.2 / 451.2 / -12691
1992 / 3 / 479.3 / -18309
1993 / 3 / 541.9 / -22777
1994 / 2.6 / 600.3 / -29505.1
1995 / 2.8 / 696.7 / -33789.5
1996 / 3 / 786.2 / -39520.2
1997 / 2.3 / 868.5 / -49695.5
1998 / 1.6 / 801.6 / -56927.4
1999 / 2.2 / 851.3 / -68677.1
2000 / 3.4 / 817.9 / -83833
2001 / 2.8 / 767.3 / -83096.1
2002 / 1.6 / 886.3 / -103065
2003 / 2.3 / 993.1 / -124068
2004 / 2.7 / 1182.6 / -161938
2005 / 3.4 / 1330.7 / -201545

Table 1

Analysis for this project consisted of the use of two SAS procedures. Firstly, PROC REG was used to perform a regression analysis of the data. Secondly, PROC CORR was used to determine correlations between the three variables. HTML tables of the results of these procedures were written by SAS and appear in this report.

III.  Results/Economic Analysis

In this trivariate analysis, the models used for the regression procedure were as follows.

·  U.S.-Chinese trade balance vs. Annual percent change in CPI

·  U.S.-Chinese trade balance vs. Corporate profits

In the former model, two iterations were used. One included “year” in the regression procedure and the other iteration did not. The correlation procedure was used to check for correlations between the trade balance and the CPI and corporate profits.

The resulting regression and correlation information displayed in Tables 2-5, was very informative.

Regression: Dependent Variable: Annual Percent Change in CPI

Number of Observations Read / 21
Number of Observations Used / 21
Analysis of Variance /
Source / DF / Sum of
Squares / Mean
Square / F Value / PrF /
Model / 1 / 2.34471 / 2.34471 / 2.57 / 0.1255
Error / 19 / 17.34481 / 0.91288
Corrected Total / 20 / 19.68952
Root MSE / 0.95545 / R-Square / 0.1191
Dependent Mean / 3.06190 / Adj R-Sq / 0.0727
Coeff Var / 31.20443
Parameter Estimates /
Variable / Label / DF / Parameter
Estimate / Standard
Error / tValue / Pr|t| /
Intercept / Intercept / 1 / 3.38586 / 0.29040 / 11.66 / <.0001
Annual__Balance / Annual U.S.-Chinese trade balance / 1 / 0.00000611 / 0.00000381 / 1.60 / 0.1255

Table 2

Regression: Dependent Variable: Annual Percent Change in CPI, analyzed with YEAR variable

Number of Observations Read / 21
Number of Observations Used / 21
Analysis of Variance /
Source / DF / Sum of
Squares / Mean
Square / F Value / PrF /
Model / 2 / 5.20431 / 2.60215 / 3.23 / 0.0631
Error / 18 / 14.48522 / 0.80473
Corrected Total / 20 / 19.68952
Root MSE / 0.89707 / R-Square / 0.2643
Dependent Mean / 3.06190 / Adj R-Sq / 0.1826
Coeff Var / 29.29777
Parameter Estimates /
Variable / Label / DF / Parameter
Estimate / Standard
Error / tValue / Pr|t| /
Intercept / Intercept / 1 / 301.55966 / 158.17733 / 1.91 / 0.0727
Annual__Balance / Annual U.S.-Chinese trade balance / 1 / -0.00000904 / 0.00000880 / -1.03 / 0.3175
Year / YEAR / 1 / -0.14986 / 0.07950 / -1.89 / 0.0757

Table 3

Regression: Dependent Variable: Corporate Profits

Number of Observations Read / 21
Number of Observations Used / 21
Analysis of Variance /
Source / DF / Sum of
Squares / Mean
Square / F Value / PrF /
Model / 1 / 1440481 / 1440481 / 188.98 / <.0001
Error / 19 / 144823 / 7622.27524
Corrected Total / 20 / 1585305
Root MSE / 87.30564 / R-Square / 0.9086
Dependent Mean / 684.30952 / Adj R-Sq / 0.9038
Coeff Var / 12.75821
Parameter Estimates /
Variable / Label / DF / Parameter
Estimate / Standard
Error / tValue / Pr|t| /
Intercept / Intercept / 1 / 430.39126 / 26.53546 / 16.22 / <.0001
Annual__Balance / Annual U.S.-Chinese trade balance / 1 / -0.00479 / 0.00034817 / -13.75 / <.0001

Table 4

Correlations for Annual Percent Change in CPI and Corporate Profits
With U.S.-Chinese Trade Balance
2 With Variables: / Corporate_Profits Annual
1 Variables: / Annual__Balance
Freq Variable: / Year
Simple Statistics /
Variable / N / Mean / Std Dev / Sum / Minimum / Maximum / Label /
Corporate_Profits / 41895 / 685.10421 / 274.88979 / 28702441 / 319.50000 / 1331 / Corporate Profits
Annual / 41895 / 3.06052 / 0.96778 / 128221 / 1.60000 / 5.40000 / Annual Percent Change in CPI
Annual__Balance / 41895 / -53202 / 54779 / -2.2289E9 / -201545 / -6.00000 / Annual U.S.-Chinese trade balance
Pearson Correlation Coefficients, N = 41895
Prob > |r| under H0: Rho=0 /
/ Annual__Balance /
Corporate_Profits
Corporate Profits / -0.95328
<.0001
Annual
Annual Percent Change in CPI / 0.34462
<.0001

Table 5

In Table 2, one can see the lack of a significant relationship between U.S.-Chinese trade balance and annual percent change in the CPI. With a p-value of 0.12, there is a relationship, but there are so many other factors affecting the CPI’s annual percent change that it is essentially insignificant. For instance, when one adds the variable YEAR into the mix in the regression operation presented in Table 3, he’ll find a p-value of 0.07 for that relationship. Clearly there are other, more significant factors at play affecting the rise and fall of the CPI from year to year.

The trade balance did, however, have a huge affect on corporate profits turned by American companies. With an Adjusted R-Squared value of 0.90 and a p-value of less than one ten-thousandth, as one can see in Table 4, there is a very strong relationship between American corporations’ profits and U.S.-Chinese trade. According to the correlation operation run on this relationship, represented in Table 5, there is a very strong negative correlation between the trade and profits. In essence, the more U.S. corporations import from China, the higher their profits rise.

In conclusion it seems that, while there are many factors affecting the CPI and corporate profits, the latter seems to be influenced heavily by the presence of large importations from China on the behalf of U.S. corporations. One could argue that there exists other, perhaps more important factors that go into the making of a prospering economy. Yet the fact remains that in a capitalist economy, the more profits corporations earn, the better it is for the economy.

Data Sources

CPI data: http://www.bls.gov

Corporate profit data: http://www.bea.gov

U.S.-China trade balance: http://www.census.gov/foreign-trade/balance/c5700.html

[1] See “Data sources” section at the end of this report.