SME
PMI (Purchasing Managers' Index) is an economic indicator determined by surveying the opinions and feedback of purchasing managers or related practitioners, including manufacturing PMI and non-manufacturing (services) PMI, which is used to measure the level of economic activity. At present, the mainstream PMI publishers in the US market are Standard & Poor's (Markit) and ISM.
There are differences in the coverage of the PMIs released by the two institutions, the survey subjects, the sample size, and the calculation weight. The S&P PMI survey respondents do not include government departments and require the surveyed companies to report within the scope of their domestic operations, focusing more on the corporate sector and reflecting the domestic situation in the United States, and the data is more stable, while the ISM PMI survey respondents include all sectors and do not restrict multinational companies from reporting their overseas operations, reflecting the overall changes in the United States under the international environment and more volatile.
In this article, we focus on the relationship between the manufacturing PMIs released by the two institutions. We find that the manufacturing PMI differential can characterize the fundamental differences between the relative and non-US of the United States to a certain extent. There is a certain positive correlation between the year-on-year GDP growth difference between the United States and Africa, the dollar index, the COMEX copper-LME copper price ratio, the relative return of U.S. stocks to non-U.S. stock markets, and the U.S. relative non-U.S. interest rate differential.
1. Introduction to PMI
PMI (Purchasing Managers' Index) is an economic indicator determined by surveying the opinions and feedback of purchasing managers or related practitioners, including manufacturing PMI and non-manufacturing (services) PMI, which is used to measure the level of economic activity. At present, the mainstream PMI publishers in the U.S. market are Standard & Poor's (i.e., Markit) and ISM.
The manufacturing PMI, which measures the level of economic activity in the manufacturing sector, covers output, new orders, employment, the speed of supplier deliveries, and inventories. When calculating the headline PMI, only five sub-data are calculated: new orders, output, employment, supplier deliveries, and inventories. The five data of customer inventory, prices, order backlog, new export orders, and imports are used as references. When the manufacturing PMI is above 50, it indicates an expansion in economic activity, and when it is below 50, it indicates a contraction in economic activity.
The ISM Non-Manufacturing PMI, which measures the level of economic activity in the non-manufacturing sector (e.g., services, construction, etc.). The S&P Services PMI only includes services and excludes non-manufacturing sectors such as construction, covering a smaller range of sectors. Metrics such as new orders, business activity, employment, supplier delivery speed, and inventory are also included. When calculating the headline PMI, only four sub-data are calculated: orders, business activities, employment, and supplier delivery speed. Inventories, prices, order backlogs, new export orders, imports, and inventory sentiment are six data as references. Similarly, when the non-manufacturing or services PMI is above 50, it indicates an expansion in economic activity, and when it is below 50, it indicates a contraction in economic activity.
The PMIs released by the S&P and ISM are close in the general cycle, but there are clear divergences in some periods, which may be related to differences in their coverage, survey respondents, sample size, calculation weights, etc. In contrast, the S&P PMI survey respondents do not include government departments and require the surveyed companies to report within the scope of their domestic business in the United States, so they pay more attention to the corporate sector and reflect the local situation in the United States, and the data is more stable, while the ISM PMI survey subjects include all departments and do not restrict multinational companies from reporting their overseas business, which reflects the overall changes in the United States under the international environment and is more volatile. In the calculation method, the S&P PMI pays more attention to changes in demand, with a high weight of new orders and output, and a low weight of supplier delivery and inventory, with smaller long-term fluctuations and more stability, while the ISM PMI reflects more comprehensive changes and the current situation of the whole chain, and is more sensitive to marginal changes.
From a correlation perspective, the correlation between the S&P manufacturing PMI and the year-over-year GDP growth rate in the United States is significantly stronger than the ISM manufacturing PMI in most years. The correlation between the ISM manufacturing PMI and the year-on-year growth rate of global GDP will be slightly stronger than that of the S&P manufacturing PMI in most years, but the correlation between the S&P manufacturing PMI and global GDP has increased significantly after the epidemic, which may be related to the reshaping of the global industrial chain and the rebound in the proportion of the U.S. manufacturing industry.
The non-manufacturing sector is the backbone of the U.S. economy, and typically the non-manufacturing sector guides trends in changes in economic activity, while the manufacturing sector provides volatility. The overall fluctuation of the non-manufacturing PMI is smaller than that of the manufacturing PMI, and the cyclical nature is weaker, especially after the subprime mortgage crisis, the characteristics of its cyclical fluctuations have further declined. The prosperity of manufacturing PMI is more closely related to the rotation of the inventory cycle and the changes in the prices of major assets, and the correlation is stronger. At the same time, the data of the ISM manufacturing PMI began in 1948, and the data sample available for study is more sufficient. Therefore, in this article, we mainly select manufacturing PMI for analysis.
Historically, the manufacturing PMI new orders sub-item has a leading effect of about 4 months on new orders for durable goods, and the current new orders sub-item continues to rebound, and new orders for durable goods will not be ruled out in the future. The price sub-item has a 5-month leading effect on the CPI, and at present, the price sub-item has signs of stabilizing and rebounding, and the downward slope of the CPI may slow down or even reflation in the future. Compared with the unemployment rate and non-farm payrolls, the employment sub-item of PMI has the same general trend, but the short-term fluctuation of the employment sub-item is greater. The PMI employment sub-item has a leading effect on the year-on-year growth rate of non-farm employment for about 4 months. Comparing the PMI supplier delivery sub-item with the supply chain index, the two were close in general trend before 2020, but often diverged, but after 2020, the performance of the two was almost synchronous.
Second, the correlation between the manufacturing PMI difference between the two institutions and economic fundamentals
Our research finds that the difference between the two manufacturing PMIs can depict a portion of the U.S. economic relative advantage over non-U.S. economies.
First, we calculate the difference between the S&P manufacturing PMI and the ISM manufacturing PMI as:
Since the S&P manufacturing PMI mainly covers the domestic data, the data performance of the domestic economy in the United States is usually better than that of the non-US economy. The GDP index of non-US advanced economies according to the Dallas Fed represents the growth of non-US economies year-on-year. There is a positive correlation between the difference between the manufacturing PMI and the year-on-year GDP growth difference between the United States and non-American developed economies, and there is a leading effect of about two quarters. Judging from the current level of S&P manufacturing PMI vs ISM manufacturing PMI, the United States still has a certain advantage over the non-American economy, but the margin has converged.
3. The correlation between the difference between the manufacturing PMI and asset prices of the two institutions
3.1 U.S. Dollar Index
The U.S. dollar index mainly measures the strength of the U.S. dollar against a basket of developed market currencies, and its trend is often affected by the relative performance of the economy and the relative tightening (easing) of monetary policy. Since the manufacturing PMI differential can depict the difference in growth between the United States and non-US economies, it has a positive correlation with the U.S. dollar index. In addition, the manufacturing margin has a three-month lead over the U.S. dollar index, so it seems that the U.S. dollar index will remain relatively strong in the short term.
3.2 Copper
During the period when the S&P manufacturing PMI is strong relative to the ISM manufacturing PMI, the COMEX copper price will also show a premium to the LME copper price. This may be because copper is one of the bellwethers of the manufacturing industry, and during the period when the S&P manufacturing PMI is strong, the domestic manufacturing industry in the United States has stronger demand than overseas manufacturing, which makes COMEX copper appear at a premium.
3.3 U.S. stocks
The Dow Jones Industrial Index is made up of 30 large and significant companies listed on major U.S. exchanges, with more stable earnings from its constituents. Nasdaq constituents tend to be high-tech, biotech, and other innovative companies with a higher proportion of technology stocks that are more affected by valuations. The S&P 500 index includes the 500 largest publicly traded companies in the United States. From the perspective of the correlation coefficient matrix, the highest correlation is between the ISM manufacturing PMI and the Dow Jones Industrial Index. This may indicate that the ISM manufacturing PMI indicator is more accurate in describing the prosperity of the traditional economic sectors in the United States.
Compare the difference between the manufacturing PMI spread and the annual return of U.S. stocks against non-U.S. developed economies (both U.S. stocks and non-U.S. developed economies choose the MSCI index as a reference). It was found that there was a certain positive correlation for most periods. The trend of the stock market integrates factors such as economic fundamentals, monetary policy, sovereign credit and other dimensions, and the manufacturing PMI difference mainly depicts the difference between the fundamentals of the United States and non-US economies, which has a more significant impact during the period when fundamental factors are dominant.
3.4 Treasury yields
The correlation between the manufacturing PMI itself and Treasury yields is unstable. There is a positive correlation during periods of low interest rates, and negative correlations during periods of rapid rate hikes. The correlation between the S&P manufacturing PMI vs ISM manufacturing PMI and U.S. Treasury yields also changes cyclically, with a negative correlation before 2020 and a positive correlation after 2020.
Compare the manufacturing PMI differential with the U.S. interest rate differential against non-U.S. economies. Among them, the U.S. yield is the 10Y U.S. Treasury yield, and the non-U.S. economy interest rate differential is the Dallas Fed's long-term interest rate for non-U.S. developed economies. From 2007 to the present, there has been a positive correlation between the two, but there have been periodic divergences. In general, when the monetary cycles of the U.S. and non-U.S. economies are more synchronized, when the U.S. has a greater advantage over the non-U.S. economy, the interest rate differential between the U.S. and the non-U.S. economy will be larger, so most of the periods show a positive correlation. For example, from the second half of 2018 to 2019, when the market began to expect the Fed to cut interest rates, the U.S. interest rate differential towards non-U.S. interest rates converged, but the U.S. economy still maintained a relatively better performance, and the interest rate differential and the PMI spread would be negatively correlated. In 2016 and 2017, the U.S. economy was relatively weak, but the Federal Reserve has started to raise interest rates, and most of the central banks of other developed economies have maintained low interest rates.
The historical correlation between the manufacturing PMI spread and the US-China interest rate spread (10Y U.S. Treasury yield - 10Y China bond yield spread) is not strong, probably because the economic growth rate and interest rate differential between China and the United States are large, and the market is not sensitive to this. After 2020, as the gap between China's and the United States' economic growth rate narrows and the U.S.-China interest rate gap gradually converges, the manufacturing PMI difference and the U.S.-China interest rate differential begin to show a high positive correlation. If we compare the difference between the S&P US manufacturing PMI and the official Chinese manufacturing PMI and the US-China interest rate differential, the two have begun to show a strong positive correlation after 2016.
★
★
Disclaimer
CIB Research Co.,Ltd. Chinese The information provided in this report is obtained in accordance with international and industry standards and through legal channels, but does not guarantee the accuracy and completeness of the information stated in the report, and readers of the report should not rely on the accuracy and completeness of the information.
The information provided in this report reflects the judgment of this report at the time of the initial public release, and we reserve the right to supplement, correct and revise the relevant information at any time, but do not guarantee that it will be released in a timely manner. The content of this report is for the reference of the report reader only, and all business decisions will be made by the report reader after synthesizing the information of all parties, and our company does not assume any responsibility for any direct or indirect consequences caused by the information provided in this report.
The judgments in this report are based on the researcher's own knowledge and inclination, and should be regarded as the researcher's personal opinion and do not represent the institution. Our company may issue other reports that are inconsistent with the information provided in this report or express different views based on objective circumstances or different data sources or analysis. The researcher himself believes that he adheres to an objective and neutral position, but does not make any guarantee that the relevant information expressed in the report is directly or indirectly related to the business interests of our company, and the relevant risks must be independently assessed by the reader of the report, and our company and the researcher do not assume any legal responsibility that may arise therefrom.
The information and opinions expressed in this report do not constitute any offer or investment advice and should not be relied upon as the basis for any investment research decision, and no action is taken to ensure that the information contained in this report is appropriate for individual investors or any individual, and we do not recommend any action based on this report.
Nor should the reader of the report interpret any statements in the report from a strictly economic perspective and without any ethical, political or other biases, and neither we nor the researchers themselves accept any responsibility for the consequences of any interpretation based on these biases, and reserve all rights to take action to protect their rights.
The copyright of this report is only owned by our company, and no institution or individual may reproduce, reproduce and publish it in any form without written permission. Unless it has been officially published in a public publication, it should be regarded as a non-public act of discussion and analysis. If it is quoted or published, the source must be indicated as "Industrial Economic Research & Consulting Co., Ltd.", and this report shall not be quoted, abridged or modified contrary to the original intention.
Our company has the right to modify and interpret the terms of this disclaimer.