The Setup:

The two big themes for China so far this year are the slowly evolving train wreck in real estate (previous coverage here, here, and here) and the Chinese consumer (coverage here and here). There is much bullish chatter about the consumer in the context of China’s reopening — whether or not the bounce will ultimately fizzle is up for debate (e.g. see our take on the real driver of household savings here). For the purposes of today’s issue, these conversations made us think about quantifying (1) media sentiment on China’s economy and (2) US-listed company China exposures. Data files attached.

Contents:

1. Diverging East-West Media Tone on China’s Economy

2. China Exposures by US-Listed Companies

3. Data Files (.csv)

1. Diverging East-West Media Tone on China’s Economy

A Global Database of Society

Supported by Google Jigsaw, the GDELT Project monitors the world's broadcast, print, and web news from nearly every corner of every country in over 100 languages and identifies the people, locations, organizations, themes, sources, emotions, counts, quotes, images and events driving our global society every second of every day, creating a free open platform for computing on the entire world.

— GDELT project

It’s actually crazy to see how humanity’s computing power allows for this level of media data collection, aggregation, and processing, all in near real-time.

The chart above shows media coverage of China’s economy (keywords: ‘China’ & ‘economy’) across print, internet, and televised media from 2017. We segment the coverage by country and language of origin — English in the Five Eyes (US, UK, Canada, Australia, New Zealand) and Chinese in Greater China. Coverage tone (top panel) is calculated by GDELT’s algorithm — high values denote positive tone, and negative values denote negative tone. Coverage volume is shown in the lower panel.

Some observations. Western media is consistently negative about China’s economy, whereas Chinese media is consistently positive (except during Covid onset at the start of 2020). Western media seem to have slid in tone at the end of 2021 (Delta wave in China?) and have recovered only slightly since. Chinese media have been slowly hyping up the economy since the beginning of 2022 (both tone and volume are in uptrend).

In terms of volume, Western media shows occasional bursts of coverage (spikes), presumably when it latches onto a story, otherwise volumes are rather low. China’s volumes clearly show difference before/after 2021, presumably when the focus shifted away from the economy to other narratives. Only recently have volumes pertaining to economic coverage reached previous highs.

Above is the difference in tone between the two regions. Uptrends indicate increasing divergence (Chinese media more positive and/or Western media more negative). Downtrends signify a narrowing of this divergence. On average since 2017, Chinese media are about 3 tone points above their Western counterparts.

Here’s a distribution chart of media coverage by the two regions. Another way to visualise the consistently positive (China) or negative (Western) tone on China’s economy.

Keep this persistent bias in mind when reading or watching the news.

2. China Exposures by US-Listed Companies

Let’s turn to company exposure — of all the US-listed companies, which companies and industries have the largest exposure to China, good or bad? Using the SEC’s Edgar database, we run a simple word search for keywords ‘China’ or ‘Chinese’ via their API within 10-Ks filed in 2022. Above is a sample text extraction of the search for S&P Global (NYSE: SPGI). Words ‘opportunity’ and ‘growth’ are used in its 10-K wording.

Aggregating this search, we sum word hits by company. We match SIC industry codes to company CIK number with some additional wrangling. The companies with the highest word hits (nword.hits) are shown in the screenshot above. China KFC is run by Yum China, which is in third place with 853 hits. BeiGene, which specialises in cancer treatment, follows in fourth place with 705 hits. Note that most Chinese companies listed in the US are required to file 20-Fs instead of 10-Ks (something to do with domicile?). We may cover 20-Fs in another post. Today we were focussed on 10-Ks only.

Now here’s the fancy data visualisation bit. We run word hit distributions by industry code and rank order them by China exposure. Industries at the top of the chart have the least exposure, while those towards the bottom have the most. Exposure is ranked by median company to avoid potential skew from averaging. Since there are 398 SIC industry classifications, we filter out industries with less than 20 companies to make the chart more legible. The industries with the highest China exposure are semiconductors, motor vehicle parts, and laboratory analytical instruments (see very bottom of chart). Asset-backed securities, blank check companies, and crude petroleum / natural gas companies have the least China exposure (see top of chart).

For investors who are seeking China exposure through US-listed equities — or, alternatively, seeking to avoid it — perhaps this rank ordering will prove useful.

Here’s the same ridgeline chart as above but without the filtering of industries with low company counts. Due to the sheer number of SIC codes, the labels become illegible but the visualisation still stands. For details, we refer our readers to the Excel files with the data towards the end of this post.

We are a pair of finance professionals with boots on the ground in China. We love putting out China Charts and are pleased that it goes out for free to over a thousand and a half readers worldwide. But it takes a lot of work and what sustains the effort is the support of paying subscribers. If you appreciate the newsletter and can afford the subscription, please hit the button and pick one of the three options.

3. Data Files (.xlsx)

All data used to make the charts in this issue are included below for premium subscribers. As always, thank you for your support.

• Data from GDELT includes the time series used for the charts and an additional ex_China_ex_West region for both volume and tone.

  

• Data for China exposures are provided by company, industry, and overseeing governmental office levels. After cleaning the data and filtering out the NAs, there are 7,702 companies in the data file.