Have you ever wondered if SOEs get taxed? If they do get taxed, why? I knew that SOEs were owned by the state, but I had never made the leap to think about how the state benefits. If you asked me, I guess I would say through distributions... But how can shareholders, even a government shareholder, compel distribution? Well if these questions have ever taxed your mind, even if a newly promulgated tax on thought, I'd recommend reading Income Taxation of State-Owned Enterprises: Theory and Chinese Evidence by Professor Cui Wei of China University of Political Science and Law. I'm going to briefly summarize Professor Cui's description of the traditional theories on SOE taxation, and then summarize his theory of why SOEs are taxed. But first, some threshold issues.
SOEs in China, and in many countries, are subject to income taxation. In China, they are subject to the same tax as private enterprises under the EIT, though SOEs do tend to be among the most aggressive in seeking favorable tax treatment. Tax on SOEs might be conceptualized as a distribution by the SOE to its shareholder, the government. This carries one major caveat: the shareholder does not pay any tax on its constructive distribution which eliminates considering the important tax planning issue of structuring distributions to best advantage the shareholders' tax liabilities.
The first theory for SOE taxation is that "SOEs are taxed because they have mixed public and private ownership." Professor Cui does not think that this explains SOE income taxation because there is no good reason why an SOE with mixed ownership should be taxed in the same way as a private firm.
The second theory for SOE taxation is that "SOEs are taxed because the government that owns them is not the government that taxes them." In federalist countries such as China, SOEs can be subject to a dual layer of provincial and national taxation. Even though an SOE might be owned at a provincial or municipal level, part of its constructive distribution is withheld in Beijing with the rest distributed to the local governments to cover the SOEs tax liability at and below the provincial level. Professor Cui argues tha this explanation fails because there are much better ways than the income tax system for equitably distributing an SOEs profits.
The third theory for SOE taxation is that "SOEs should be taxed because that is necessary to ensure that they are competitively on an equal footing with private firms." Professor Cui notes that this view may be prevalent in developing countries. Professor Cui concedes that competitive reasons are a fair explanation for other taxes such as VAT, but "the price affected by the corporate income tax is the price of capital, and SOEs and private firms generally do not compete for the same investors, unless the SOEs are being privatized." Thus any income tax is simply a burden borne by the government because different tax rates for SOEs' profits would have no effect on SOEs' ability to attract more government capital.
Instead of relying on these theories, Professor Cui proposes that SOEs are taxed, at least in China, for the sole purpose of ensuring (and forcing) periodic distributions to the government shareholder. Professor Cui argues that for his theory to prevail, SOEs must be averse to the payment of income tax. Despite their tax paying braggadocio, Professor Cui points to substantial evidence that China SOEs are as averse as their private enterprise counterparts to paying tax, thus substantiating his theory.
All in all a fine read, and I look forward to further refinements to Professor Cui's analysis as many of the questions he leaves unanswered for lack of evidence are fielded.
17 comments:
What is most telling about this article is the fact that a respected professor at a well-known university can only theorize about how and whether SOEs are taxed.
Fascinating! Thanks for highlighting this.
What's worse is that the professor misses the whole point of why the Chinese government taxes SOE's. The author appears to be a professor in tax law, and so seems to be unfamiliar with the economics literature or the history of Chinese reform.
The problem is that if the state forces the company to issue dividends, then there is zero incentive for the company to make a profit. What's the point of making any money if the state is just going to take it. In that situation it's better for the company to *lose* money because the higher the losses the more state subsidies are needed. The policy of "no dividends" was established in the early-1980's, and was part of the idea that the government would not extract dividends, in exchange, companies would accept no government subsidies. It's only in the last few years, that the government has thought about issuing dividends, and it's a messy problem, which is why the government uses taxes.
If the state mandates a given tax rate and they keeps the tax rate standard across all the SOE's in a given industry, then lots of good things happen. The SOE's has incentives to make a profit because the SOE keeps the profits and can distribute them to managers. Tax rates are set at a high level, which means that they don't change often which means that you can plan for them. Because they are standardized, you have less ability to play patronage games or collect political rates. Finally, if you force state distributions in lieu of taxes, you are putting SOE's at a disadvantage to non-SOE firms.
BTW - My id is twofish.
Also all this is quite relevant to US SOE's. One problem the US has run into with the banks is how to prevent an economic disaster without rewarding incompetence.
Paulson tried to do this by giving all of the banks equal shares of money via TARP, but that seems to have blown up. If you don't put limits on how much the government can extract from an SOE, then you end up with the "Amtrak disease."
twofish,
The professor does examine the history of tax reform in China from 1979 through the new EIT. The original "no dividends policy" collapsed within 3 years, and the government moved to the private contracting of SOE management for a fixed distribution of profits and taxes. This system was abolished with tax reform in 1994 because of suspected widespread tax evasion by the government contractors resulting in China's 1993 fiscal crisis. The 1994 tax reform brought SOEs under the same corporate income tax rates as private companies, as VAT and turnover taxes became the largest contributors to the government kitty.
Your last paragraph is largely what the regime in China is, and how professor Cui describes it. The income tax is the periodic, predictable distribution to the government shareholder. And these distributions can be huge. Professor Cui writes that "China National Petroleum Corporation . . . boasts of having 'paid' by itself 3.6% of the national tax revenue in 2007," and the companies managed by SASAC paid almost 20% of the national tax revenue in 2008.
twofish,
Ugh, I am not looking forward to finding out how the US government proposes to extract value from its 'investments.' I suspect the subsidies, until they prove to be otherwise, of the bailout will eventually be subject to WTO dispute, unless something was/is agreed upon at high level summit meetings.
Twofish again - There is a very good political reason *why* Chinese SOE's are taxed at the same rates as private firms. One persistent argument over the last thirty years of Chinese economic reform is whether and to what extent Chinese enterprises should be privatized, with advocates of privatization arguing that private firms would be more efficient, and advocates of public firms arguing that they wouldn't be.
The political compromise is for the government not to take a stand on this issue, and to let the market decide. If you treat, public, mixed, and private enterprises differently, this would have the state either punishing SOE's or punishing private firms which would destroy the SOE/private compromise.
SOE's and private firms didn't compete for the same investors in 1995, but one of the goals of SOE and private firms compete on a equal footing. One reason for that is that Chinese corporate ownership is a zoo of corporate forms, in which it (intentionally) is pretty difficult to divide between "private" and "public." It's also the case that very few Chinese enterprises are technically state-owned enterprises. They are listed enterprises in which the state has majority or near total ownership. Chinese law is intentionally set up to create the same rules for a company that has 5% state ownership as 95% state ownership. Why should they otherwise be?
One important thing about Chinese political discourse, is that you never hear about "privatization" but also about "popular ownership." The argument in favor of privatization is that by having shares distributed among the people you end up with a system that is more "socialist" in some sense than bureaucratic ownership by the state.
One final point. Talking about the Chinese state as some abstract entity miss the point that there are a lot different actors within the state. A system in which you have dividend payments, the money will end up in the hands of different people than a system in which you have taxes. Cui asks why the government doesn't ask the holders of the dividend payments to redistribute the money, and the answer is that they'll say no.
twofish: Ugh, I am not looking forward to finding out how the US government proposes to extract value from its 'investments.'
Right now, they don't know, which is why we need to be discussing this.
twofish: I suspect the subsidies, until they prove to be otherwise, of the bailout will eventually be subject to WTO dispute, unless something was/is agreed upon at high level summit meetings.
I doubt it. The WTO restrictions on capital injections are full of holes, and in trade law, "everyone does it" is a valid excuse to do something.
However, you do bring up a good point. A tax rebate on SOE's that a foreign firm cannot get would almost certainly be illegal under WTO rules. A lot of the economic philosophy that requires that foreign and domestic firms be treated the same, also go to requiring that state owned and non-state owned firms be treated the same.
Anonymous II/Not twofish (?),
Those were the thoughts of Will/Me.
"twofish: Ugh, I am not looking forward to finding out how the US government proposes to extract value from its 'investments.'
"Right now, they don't know, which is why we need to be discussing this."
I have two ideas for how value will be extracted. 1) It won't; or 2) through federal income taxation. Number two poses a major problem: the recipients of bailout money, such as US auto companies, do not have much profit to tax, and even if/when they post losses they are not pass through entities for tax purposes and the government would have no use for loss in the event of restructuring. However, number 2 is attractive in that it suggests optimism in the government's investment for the profitable future of US auto and their return as high income tax paying legal persons.
Will Lewis: The original "no dividends policy" collapsed within 3 years, and the government moved to the private contracting of SOE management for a fixed distribution of profits and taxes.
No it didn't. The original "no dividends policy" required the company to turn over a fraction of profits. In 1987, most companies were then required to give a fixed amount rather than a fractional amount. This change impacts the way the "tax" is calculated, but it doesn't affect the fact that it is a "tax" rather than a "dividend."
Which brings up an interesting point which is that in a command economy, there is no difference between a "tax" and a "dividend." It's only when you have a market economy does it make a difference. I think the essential different between a "tax" and a "dividend" is "who decides?" which gets at the heart of the behavioral explanations for SOE taxation.
One thing that you can see this is in the evolution of payments made to the government as part of the agricultural contract-responsibility system. In 1978, it was viewed as an incentive payment to the peasant for meeting a quota. After 2000, when it was abolished, it was seen as a "land tax."
I think one constructive criticism for Cui's article is that before you develop a theoretical framework for why SOE's should be taxed, you need to develop a theoretical framework that defines what a "tax" is and how it is different from a "dividend" and also exactly what "state ownership" means.
Lewis: Number two poses a major problem: the recipients of bailout money, such as US auto companies, do not have much profit to tax.
The fun thing is that this is why no one worried that much about either dividends or taxes from Chinese SOE's until recently. There's no point in worrying about how to extract value from profits, if there are no profits.
One very important part of Chinese corporate law is that preferred shares are illegal (which makes venture capital hell). If the state owns 70% of a company, and it pays a dividend, it can legally only get 70% of the dividend, and these payments are transparent enough so that people can monitor them.
The reason for this is that this quite intentionally limits the amount of money that the Chinese state can extract from the company. You aren't going to put money in a Chinese company if you know that the next day the state is just going to take all of it.
To Above Anonymous,
Switching from taxation and profit sharing based on a formula to private management and taxation and pre-fixed share of profit is a change. Though the link in the cite is not working, if this change in tax law in 1987 was a major contributor to the fiscal crisis of 1993, then the new method of calculating SOE "tax" in 1987 was a considerable change. I mirrored the 'collapse' language of Cui, and that is melodramatic on reexamination.
"I think one constructive criticism for Cui's article is that before you develop a theoretical framework for why SOE's should be taxed, you need to develop a theoretical framework that defines what a "tax" is and how it is different from a "dividend" and also exactly what "state ownership" means."
I agree, this is a good criticism. On page 16, Cui too casually resorts to the tired cliche of "this is a uniquely Chinese way of doing things that Westerner's don't fully grasp" in characterizing "taxation as a means of forced distribution:"
"the characterization of SOE taxation as a means for forced distribution has a special resonance in the Chinese context, as it bears some resemblance to how the merit of SOE taxation was discussed there in the 1980s. In fact, while characterization of taxation as forced distribution may possess novelty for some western scholars, it might seem too familiar a narrative in the Chinese context."
Regarding the preferred shares:
"One very important part of Chinese corporate law is that preferred shares are illegal (which makes venture capital hell)."
I've been intrigued by the rumored rise of the CJV in China VC to create the legal fiction of preferred shares. A CJV agreement creating the legal fiction of preferred shares can't be anymore complicated than the Articles and Bylaws in a "typical" US VC investment, can it?
Saying that China's situation is unique isn't useful. Every economy has some unique aspects to it, and part of the goal is to create a theoretical framework that explains *how* something is unique.
One of the ironies is that Cui talks about Chinese uniqueness, but then misses the unique part about Chinese SOE's which is that they are standard industrial enterprises which perform the same functions as privately owned enterprises in the other countries or even in China. Any enterprises for which the government believes has a "public goods" function that should be funded from tax revenue, is not going to structured as a state owned enterprise in China.
It's going to be part of the state. Chinese SOE's exist precisely because the government *didn't* want them to be supported by tax revenue, thereby hitting soft budget constraints.
Another factor is that Chinese corporations have all sorts of baroque legal structure, and any difference in government treatment based on corporate structure is going to be gamed. It's often not clear under Chinese law, what is or is not a state owned enterprise, since some nominally privately owned enterprises are ultimately managed by state officials and state ownership of some form is often a cover for private ownership.
SOE's have a specific meaning under Chinese law, and most SOE's aren't SOE's under the Chinese legal definition.
Some other reasons SOE's have taxes rather than dividends has to with the fact that the actual distribution of power in a lot of companies is different than the idealized shareholder -> board -> manager system.
It's quite common for the effective state shareholder to also be the de-facto manager in a corporation, in which case they have no interest in issuing dividends which presumably would go to another unrelated part of the state, rather than to the managers pocket, which would be case if no dividend were issued. In this situation, the state would have to impose a tax to insure that the corporation disgorges anything.
Finally, I suspect that a lot of the "reorganizations" that got favorable tax treatment were actually corporate bankruptcy restructurings, in which a healthy corporations takes over the ongoing liabilities of a bankruptcy corporation in exchange for a tax exemption.
Lewis: A CJV agreement creating the legal fiction of preferred shares can't be anymore complicated than the Articles and Bylaws in a "typical" US VC investment, can it?
I think it is. Typically, a US VC investment has standard Articles of Incorporation and Bylaws, and any transactions issuing preferred stock or agreements regarding the relationship between the investors are done through shareholder agreements, contracts, and board resolutions. Putting any of this in the bylaws adds unnecessary complexity and just generates a mess.
In US corporations, the directors or common shareholders often are able to unilaterally amend and certainly block amendments the bylaws, which is not a good thing if you are a preferred shareholder, or if you don't have a majority on the board, and if you have to amend the bylaws each time you want to issue stock. Finally some states have statutory restrictions on corporate structure and bylaw provisions which are another set of land mines.
If anything has to be litigated then litigating the bylaws just adds a mess in comparison to litigating a contract. US law is pretty open about what can be in a contract.
In China, people do very clever things because they have no other options.
Anonymous,
First, sorry for not noticing your final comment in my inbox sooner.
Second, thank you for taking so much time to respond so constructively. I've learned a lot from this discourse.
If I have some spare time, I might want to write a working paper to respond to Cui. What I would find interesting is some comparative knowledge since I know something about Chinese SOE's and US corporate structure, but practically nothing about SOE's in other countries.
The basic answer to Cui is that in the Chinese system of a socialist market economy, the state is merely just another owner for the SOE's and the SOE's have the same functions as role as private commercial enterprises in other countries or as in China itself. Since the state is "just other shareholder" there is no reason for and lots of reasons against treating the state in any way that isn't "just another shareholder."
What would be interesting is to see how this works in other countries.
One point that a lot of Americans don't realize is the extent to which US corporations are owned by state entities. US state governments have huge pension funds and university endowments such as Calpers, which are used to exercise ownership of US corporations.
The similarity between US and Chinese ownership structures is not a concidence as a lot of Chinese state ownership structures such as the China Investment Corporation seem to be quite intentionally modelled on similar institutions in the US such as Calpers. The sharing goes both ways.
It seems pretty obvious to me, although no one is publicly mentioning it, that the way the US is structuring ownership of General Motors and Citigroup draws from the Chinese experience in restructuring its SOE's.
I realize that I am piling on a little late, but I have two points to add. First, you should note that the PRC treats some SOEs (like BOCOG) as branches of the government. Like provincial governors or ministers, SOE leaders assume particularly high positions in the Politburo after successful management of Chinese SOEs. Moreover, several larger Chinese SOEs have essentially co-opted PRC foreign policy in geographic areas of interest for the Chinese government. As such, it is difficult to make an apples-to-apples comparison; notwithstanding John Perkins' "Confessions of an Economic Hitman," PRC SOEs in so called "strategic industries" have not and never will behave solely as vehicles for investment, like US corporations are at least intended to act.
Second, if you're looking for some good comparative sources on SOEs, and if I can give myself a shameless plug, take a gander at my two papers on Chinese SOEs. The first is located here and the second is located here. The citations in both works could be a good starting point for subsequent research, and the one paper on international law gets into some of the issues I described above in greater detail.
Post a Comment