Paul
The World Bank World Development Indicators have a series on China. However, most of the datasets are from 2013. That is, they lack any real longitudinality (even if such a word exists).
The best that I can see in those data sets is "Taxes on income, profits and capital gains (current LCU) - China". This might be a proxy for surplus value if the Chinese Government taxes on the basis of Marxian economics. However, there is no cyclical element in the data (here on Quandl: https://www.quandl.com/data/WWDI/CHN_GC_TAX_YPKG_CN-Taxes-on-income-profits-and-capital-gains-current-LCU-China and from here on the WB web site: http://data.worldbank.org/country/china#cp_gep ).
I hope this helps
Rob
hi Paul: as Rob says, this is difficult as the required time series are relatively short.
More importantly, definitions have changed markedly, as the economy has changed. Some data in the NBS files are for medium and large firms; some are for all. Some are for manufacturing, and some more general. And so on.
In my 2012 book, Making capitalism in rural China, I made an attempt to estimate the history of the rate of surplus value in one region [Suzhou, in Jiangsu] and if you look at chapter 3 you can see some of the convolutions I had to go through. I think that the results were interesting, though, so please do persevere.
Michael
There might be relevant data at the link below, but you may need to use an alternative equation for calculating surplus value. Part of the difficulty here is the distinction between measuring s/v relative to labor time, which in part is how Marx discussed surplus value. In other examples he gave monetary value terms (as transformations of labor into price). While I cannot claim to be an expert on this issue I have used similar data. Part of the problem, which we see even in Marx, is one related to measurement of s/v and data availability, which lead Marx to use sometimes abstract examples for labor time examples (ie., where he uses labor time there is no indication that these were real labor times versus data he simply employed to illustrate his point), and sometimes the use of "real world" or monetary measures in other examples (e.g., converting s and v into monetary terms).
In any event, this website -- https://www.ceicdata.com/en/countries/china
suggests it has some economic indicators for China beginning in 1949. I do not know the quality of the data (it is a fee for data service), but the website indicates you can get a free trail period.
https://www.ceicdata.com/en/free-trial-ccd
I did series for a decade or so using the Chia statistical yearbook. The big problem with that is that you had no capital stock figures. I had to get them from investment figures plus a depreciation model. It was made worse by having to aggregate regional figures.