Estimating PRC's Round Tripping FDI

According to PRC's official definition, Foreign Direct Investment (FDI) refers to the investment in three legal types of foreign invested enterprises (FIEs) in PRC: solely foreign funded enterprises, sino-foreign joint ventures and sino-foreign cooperative ventures. The foreign investors in FIEs include any foreign enterprise, economic entity or individual as well as the Hong Kong, China, Macao and Taipei,China compatriots and the Chinese enterprises registered outside PRC. FDI must be invested in the form of spot foreign exchange, in-kind, or technology investment. The re-investment of the profits by FIEs and the funds borrowed from overseas by the FIEs for their PRC projects can also be counted as FDI.

Round-tripping FDI refers to the domestic capital that has fled the home country and then flows back in the form of foreign direct investment. In the case of PRC, it could also include domestic capital that is counted as foreign capital against the government regulation. This often happens to the foreign invested component of the registered capital for a newly established foreign invested enterprise. The faking of the foreign invested component of the registered capital could involve PRC's commercial bank lending to the foreign invested enterprises in violation of PRC's relevant regulations. It is common for some fake foreign invested enterprises to use false capital auditing report and false bank deposits documents to meet the requirements of registered capital input by the foreign partners. These incidences would clearly inflate the FDI statistics reported by the Chinese authorities.

The inflated FDI inflow statistics as reported by PRC will be much higher than the FDI outflow statistics as reported by the source region since there are no incentives for foreign investors to report their fake investment in PRC to their home countries. Hence, the gap between FDI inflow statistics as reported by PRC and FDI outflow statistics as reported by source regions are the unverifiable or unconfirmed part of PRC's FDI inflows and can be used as a proxy measure of the round tripping FDI to PRC. This is the methods used in this paper to estimate PRC's round tripping FDI from the Hong Kong, China and other source regions.

Round Tripping FDI from U.S., Germany, Japan, Republic of Korea, Taipei,China Province, and Singapore

In this sub-section, we try to estimate the round tripping FDI to PRC from six source regions which has published their own independent statistics on FDI to PRC. The round tripping FDI from Hong Kong, China will be discussed in the next sub-section as the case of Hong Kong, China is more complicated than other source regions. Table 10.1 to 10.6 [ PDF 138KB | 7 pages ] shows FDI statistics as reported by PRC and the source regions, including U.S., Germany, Japan, Republic of Korea, Taipei,China Province, and Singapore:

• Row A of Table 10.1 to 10.6 [ PDF 138KB | 7 pages ] is the FDI flows from the source region to PRC as reported by the source region.
• Row B is the FDI flows from the source region to PRC as reported by PRC.
• Row C is equal to Row B minus Row A and is the unverifiable FDI flows from the source region to PRC. Part of Row C is likely caused by round tripping FDI.
• Row D is the ratio of Row C over Row B, which is the ratio of unverifiable FDI.
• The last column of Row D is the weighted average of Row D over the recent years, which should have two components: the round tripping FDI ratio and the percentage of FDI that can be accounted by statistics reporting errors.
• Since Row D (the ratio of unverifiable FDI) fluctuated over the years, one column in the table also shows the standard deviation of Row D over the recent years.

It is useful to point out a few issues about the statistical reporting errors. They are related to many of the inconsistencies between PRC's and source regions' FDI statistics reporting practices. Many factors, in addition to round tripping FDI, such as the differences in the definition and collection of the FDI statistics across countries, may contribute to the above unverifiable part of PRC's FDI from each of the source regions. The appendix in an OECD Investment Policy Review "China: Progress and Reform Challenges" (OECD 2003b) provides a detailed comparison on these differences and some of the relevant parts are summarized here:

• PRC does not put a limit on the percentage of shares owned by investors (for example above 10% under OECD standards) when calculating the FDI statistics. So any amount of investment into the foreign invested enterprises by an individual or firm are considered FDI. This would inflate PRC's FDI inflows as reported by PRC relative to corresponding FDI outflows as reported by OECD countries. But the gap caused by this should not be counted as round tripping FDI. Instead, it should be regarded as one kind of the statistics reporting errors.
• The local government department in charge of FDI promotion is responsible for collecting and reporting FDI statistics, leading to serious conflict of interest and a tendency for PRC's FDI inflows as reported by PRC to be high than the FDI outflows as reported by the source region. This part can be counted as round tripping FDI.
• PRC only reports statistics on FDI inflows and does not report the statistics on market value of FDI stock, FDI outflows and incomes derived from FDI. These have made it difficult to cross-check the reliability of PRC's FDI inflows. It means that the method we are using to estimate PRC's round tripping FDI has a wide range of errors and should be interpreted accordingly.

Clearly some of the unverifiable FDI from source regions to PRC (Row C in Table 10.1 to 10.6 [ PDF 138KB | 7 pages ]) are not round tripping FDI. In another word, it seems justifiable to interpret the unverifiable FDI inflows as the high estimation or the upper bound of PRC’s round tripping FDI. On the other hand, the inconsistent accounting framework discussed above by OECD study is not entirely statistical errors. The systematic accounting bias could be regarded as over-reporting on the PRC side which is similar to round tripping FDI in nature. Also, the real statistical reporting error should have bias in both directions. By looking at Table 10.1 to 10.6 [ PDF 138KB | 7 pages ], we can see that the unverifiable part of FDI into PRC is mostly positive and large. This implies that the round tripping component of the unverifiable FDI is probably dominating the unbiased statistical reporting errors component.

To explore the problem of statistical reporting errors further, we attempted in Table 10.1 to 10.6 [ PDF 138KB | 7 pages ] to estimate the unverifiable FDI for the U.S., using the same method as we used in Table 10.7 [ PDF 63.1KB | 1 page ] for the available data from eight countries: Mexico, Brazil, Finland, Canada, Hong Kong, China, U.K., Japan, and Germany. The results are quite illustrative. Mexico and Brazil reported the same FDI statistics as U.S. so that the unverifiable FDI to U.S. from the two countries are zero. The unverifiable FDI to U.S. from Finland and Canada are very small at about 4% level. The unverifiable FDI to U.S. from Hong Kong, China and U.K. are very large at 44% and 55% levels respectively, which are close to those observed in FDI to PRC. However, the difference is that the unverifiable FDI to U.S. from Japan and Germany are negative and large at -83% and -104% levels respectively. This means that the unverifiable FDI to U.S. are more likely due to statistical reporting errors. Indeed, the weighted average of the ratios of unverifiable FDI across different source countries is small at the level of 18%. We can draw two important implications from the U.S. case:

• If the unverifiable FDI is mainly due to statistical errors like in the case of U.S., it should show both positive and negative errors. In the case of PRC, we have observed consistently large positive unverifiable FDI inflows for all of the source regions where data are available. The available source regions accounted for 70% of PRC's FDI inflows. Hence, the unverifiable FDI inflows in the PRC case is mostly likely due to round tripping FDI, instead of statistical errors.
• The statistical errors of FDI data could be huge and the unverifiable FDI could reach 50% to 100% for some source countries. Hence, our method of comparing the reported FDI statistics from host and source countries to estimate the round tripping FDI should allow large margins of error.

Now let's go back to look at the results of Table 10.1 to 10.6 [ PDF 138KB | 7 pages ], the weighted average of the unverifiable FDI to PRC in recent years is all positive and high for the six FDI source regions where data are available, e.g. 68.5% from U.S., 31% from Germany, 60.9% from Japan, 60.3% from Republic of Korea, and 70.2% from Taipei,China Province, and 65.5% from Singapore. How to interpret these numbers?

• If there are no significant statistical reporting errors, these numbers here could be regarded as a proxy for the average ratio of round tripping FDI into PRC.
• If the statistical reporting errors are non-biased in the sense that they average to zero over the years, the above numbers would also be regarded as a close approximation of the round tripping FDI ratio.
• If the statistical reporting errors have systematic bias towards over-reporting by PRC side independent of the round tripping bias, we should then adjust the above numbers downward by the size of the systematic statistical reporting errors.

To be conservative, we will use the last interpretation and to allow some systematic statistical reporting errors that are biased toward the same direction as the round tripping FDI bias. How to decide the size of the adjustment? As can be seen from Row D of Table 10.1 to 10.6 [ PDF 138KB | 7 pages ], there are large fluctuations in the unverifiable part of FDI in PRC for each of the six source regions. The degree of variation in the unverifiable part of FDI (Row C) over time is captured by its standard deviation. The standard deviation is a useful indicator on the likely range of both statistics reporting errors and the volatility of round tripping FDI. We do not have enough information to distinguish how much of the standard deviation is attributable to each of the two factors. Hence, we assume the systematically biased statistics reporting errors is as large as one half of the standard deviation of Row C (the unverifiable part of FDI) during the observed period. This is a strong assumption but is a conservative assumption for estimating round tripping FDI. We can then subtract one half of the standard deviation from the weighted average of unverifiable FDI (Row D of last column) to get the mean or middle estimate for the round tripping FDI ratio. We also use a band of errors of one half of the standard deviation to get the high and low estimates of the round tripping FDI ratio.

As shown in Table 10.1 to 10.6 [ PDF 138KB | 7 pages ], the one standard deviation for unverifiable FDI (Row D) is 13.5% for U.S., 17.2% for Germany, 18.3% for Japan, 23% for Republic of Korea, 36.2% for Taipei,China Province, and 11.2% for Singapore. The mean estimate of round tripping FDI and the associated range of errors is then:

• 61.8% for U.S. (or in the range of 55.1% to 68.5%);
• 22.4% for Germany (or in the range of 13.8% to 31%);
• 51.7% for Japan (or in the range of 42.6% to 60.9%);
• 48.8% for Republic of Korea (or in the range of 37.3% to 60.3%);
• 52.1% for Taipei,China Province (or in the range of 34% to 70.2%)
• 59.9% for Singapore (or in the range of 54.3% to 65.5%).

Round Tripping FDI from Hong Kong, China

In recent years, a rising proportion of Hong Kong, China's outward FDI is towards the Mainland PRC, 41.1% in 1998, 52.3% in 1999, 78.1% in 2000, 74.9% in 2001, and 91.3% in 2002. By comparing the Hong Kong, China's and PRC's FDI statistics we can derive the pattern of round tripping FDI from Hong Kong, China. We can use the same method as applied to the other six source regions to estimate the ratio of round tripping FDI from Hong Kong, China to PRC. But unlike the above cases, Hong Kong, China is a major international financial centre for PRC. In particular, many Chinese companies have been listed in Hong Kong, China’s stock markets. This has important implications for estimating the round tripping FDI form Hong Kong, China to PRC. Hence, we will review briefly the background of capital market development related to PRC and Hong Kong, China.

PRC made little progress in attracting foreign portfolio investment during 1997 to 2001. According to IMF 2003, the derived amount of foreign portfolio investment in PRC increased only slightly from $19.3 to $20.1 during this period, reflecting its stagnant B shares market, which is a tiny experimental stock market designed for foreign investors with share prices quoted and traded in foreign exchange according. But it was well known that even before PRC opened its B share markets to its own residents, many shareholders of B shares were actually Chinese residents using borrowed foreign passports and foreign bank accounts to carry out transactions. This is also a kind of round tripping capital flows but in the form of portfolio investments.

In March 2001, PRC opened its "B share" market to domestic residents with foreign exchange savings. This opening caused a brief surge in prices and many foreign investors took profits and dumped many shares to domestic residents. At the end of 2002, PRC announced its plan to allow the Qualified Foreign Institutional Investors (QFII) to invest in its "A share" market designed for domestic investors with RMB savings. The Chinese authorities are also studying actively the mechanism of Qualified Domestic Institutional Investors (QDII), which would allow Chinese residents to invest in overseas securities markets, including Hong Kong, China markets, where many Chinese companies are listed but their shares cannot be sold to Chinese residents through legal channels. When the cross-border transactions in the capital markets are possible, more round tripping capital flows would happen legitimately. But even before the QDII is allowed officially, many Chinese residents are already using their flight capital to buy Hong Kong, China stocks, including IPOs of Mainland companies listed in Hong Kong, China. This kind of round tripping capital flows is looking for better risk adjusted return in Hong Kong, China’s markets than in the Mainland capital markets. They will not usually be classified as round tripping FDI as the investors’ share in one listed company is usually well below 10%, the threshold for qualifying as FDI.

However, the IPO of large Mainland companies may lead to large round tripping FDI. The process is similar to the mergers and acquisitions. When a Mainland company is preparing for listing in Hong Kong, China as a "Red Chip" company, it would register as a new local company in Hong Kong, China but with a huge injection of capital from its Mainland parent company in the form of buying up a large trunk of the shares in the Hong Kong, China "Red Chip" company (usually about 60% to 70%). This would count as FDI from the Mainland to Hong Kong, China since the portfolio investment exceeds the 10% threshold for qualifying as FDI according to Hong Kong, China statistical reporting practices. Hence, the listing of Mainland PRC companies in Hong Kong, China would lead to a large FDI inflow from PRC to Hong Kong, China.

The "Red Chip" company located in Hong Kong, China then can use the capital injection from its parent company in Mainland PRC and the funds being raised from IPO in Hong Kong, China to buy substantive profit-generating projects in PRC, perhaps from some related companies under the supervision of the "Red Chip" company’s parent. This again would count as FDI from Hong Kong, China to PRC according to international practice since the procurement of projects in PRC by Hong Kong, China listed "Red Chip" companies are usually more than the 10% threshold for FDI investment. Hence, the listing of Mainland PRC companies in Hong Kong, China would lead to a large FDI inflow from Hong Kong, China to PRC. The complication here is that according to PRC’s current FDI reporting practices, the FDI investment resulting from listing Mainland companies in Hong Kong, China is not counted in PRC's FDI statistics since there is little movement of physical capital or cash in the process.

Indeed, in reality not much net capital has been moved across the border. Instead, only the ownership structure has been changed significantly and the value of the listed company may have increased a lot due to expectations about better profitability and better corporate governance. This would be the type of round tripping FDI that is intended to get value added financial services from Hong Kong, China. This type of round tripping FDI is similar to the M&A related FDI in the developed economy.

The significance of this type of round tripping FDI into PRC can be seen from the structure of Hong Kong, China’s capital markets. The share of Hong Kong, China’s market capitalisation by the Mainland background companies increased from only 4.8% in 1992 to 16.3% in 1997, 21.1% in 1999 and 26.3% in 2002. Table 12 [ PDF 58.6KB | 1 page ] shows that the share of IPO funds raised by the Mainland background companies listed in Hong Kong, China has increased from around 30% in 1991 to around 84% in 2002. Table 12 [ PDF 58.6KB | 1 page ] shows the top 10 IPOs in Hong Kong, China over the period from 1997 to 2002. Clearly, Hong Kong, China stock markets are very active in listing Mainland companies. This means that there must be significant round tripping FDI between Hong Kong, China and PRC with the purpose of using Hong Kong, China's value added capital market services.

However, it is difficult to estimate this sort of round tripping FDI since PRC does not count the financial transactions through the stock markets as FDI even if the investment is more than 10% of the companies' equity. In many IPO cases involving Hong Kong, China stock markets, no physically new foreign invested enterprises are established in PRC and little net foreign exchange capital is invested in PRC. But the impact of this sort of round tripping FDI related to capital market transactions is very significant to Hong Kong, China FDI statistics, especially in 2000. As shown in Table 13 [ PDF 70.2KB | 1 page ], in 2000, Hong Kong, China recorded $46.3 billion FDI to PRC but PRC only reported $15.4 billion FDI from Hong Kong, China. This is contrary to the general pattern during the period of 1998-2002 (except 2000) when the FDI flows from Hong Kong, China to PRC as reported by PRC were always larger than the FDI flows from Hong Kong, China to PRC as reported by Hong Kong, China. The difference between $46.3 billion and $15.4 billion is as large as $30.9 billion and can only be explained by round tripping FDI related to IPOs activities in Hong Kong, China by Mainland companies.

Indeed, as shown in Table 12 [ PDF 58.6KB | 1 page ], three of the top ten IPOs in Hong Kong, China for the period of 1997-2002 (e.g. China Unicom, Sinopec, and Petro China) were carried out in the year 2000 by large Mainland companies. The three Mainland companies raised about $12 billion through IPOs in Hong Kong, China stock markets in 2000. The IPO value of PRC's large companies is usually much smaller than one third of their total market capitalization due to large non-tradable shares by the state agencies. Hence, the parents of the above three newly listed companies must have held non-tradable shares exceeding $24 billion. Clearly some of the capital market transactions relating to these IPOs are included in Hong Kong, China's FDI statistics but not included in PRC’s FDI statistics. It is not clear how exactly Hong Kong, China companies have treated these transactions when they reported their FDI statistics. By examining the sector statistics, we found that the surge in 2000 in Hong Kong, China's FDI flows to PRC is concentrated only in the communications sector. As shown in Row A4 and B2 in Table 13 [ PDF 70.2KB | 1 page ], in 2000, Hong Kong, China reported $33.2 FDI outflows to PRC in the communications sector but PRC only reported $1.0 billion FDI inflows from all sources into the transportation, storage, post, and telecommunication services sector. Clearly much of the surge in Hong Kong, China's FDI to PRC in the year 2000 can be explained by the FDI flows in the communications sector.

Table 13 [ PDF 70.2KB | 1 page ] provides three versions of FDI flows from Hong Kong, China to PRC as reported by Hong Kong, China (e.g. A1, A2, and A3). A1 is the unadjusted FDI from Hong Kong, China to PRC. A2 is FDI from Hong Kong, China to PRC adjusted by simply excluding FDI from the communications sector (A2=A1-A4). After this adjustment, FDI from Hong Kong, China fell in all years during 1998-2002. The downward adjustment is particularly sharp for the year 2000, falling from $46.3 billion to $13.1 billion. This simple adjustment would exclude some of the regular FDI in the communications sector that are not related to capital market transactions. A3 is the FDI from Hong Kong, China with a less dramatic adjustment that allows for the regular FDI from the communications sector but excludes the apparent over-reporting by Hong Kong, China in the communications sector (A3 = A1- (A4-B2)). In A3, only the difference between A4 (FDI outflows from Hong Kong, China to PRC in the communications sector) and B2 (FDI inflows to PRC in the transportation, storage, post, and telecommunications services sector) are subtracted from the unadjusted FDI from Hong Kong, China to PRC (A1).

The FDI from Hong Kong, China to PRC as reported by Hong Kong, China and adjusted for the over-reporting by Hong Kong, China in the communications sector (A3) is compared with B1, which is the FDI from Hong Kong, China to PRC as reported by PRC. Using the same method as in the cases of the six FDI source regions, Row C in Table 13 [ PDF 70.2KB | 1 page ] (C=B1–A3) is the unverifiable part of FDI from Hong Kong, China to PRC. Row D (D=(B1-A3)/B1) is then the ratio of unverifiable part of FDI from Hong Kong, China. Following the method in the last sub-section, the weighed average of Row D can be used as the high or upper bound estimate on the ratio of round tripping FDI from Hong Kong, China to PRC. Clearly Row D fluctuates from as high as about 70% in 1998, 2001, and 2002 to as low as 8.3% in 2000. The weighted average of Row D is 53.4%. The standard deviation for Row D during 1998- 2002 is 27.1%. As in the previous cases, we will use the one half of the standard deviation as a proxy for the systematically biased statistics reporting errors. Subtracting one half of 27.1% from 53.4%, we obtain the middle or mean estimate of the round tripping FDI from Hong Kong, China to PRC, which is 39.9%. In another word, based on the available FDI statistics from Hong Kong, China and PRC, the ratio of round tripping FDI from Hong Kong, China to PRC during the period of 1998- 2002 is likely to be in the range of 26.3% to 53.4% with the middle estimate at 39.9%. It should be noted that this estimate of round tripping FDI from Hong Kong, China to PRC includes only the type of round tripping that is related to escaping regulations and does not include the type of round tripping that is related to capital market transactions such as listing Mainland companies in Hong Kong, China's stock exchange.

Round Tripping FDI from Offshore Centres

We have pointed out in the previous section that the offshore financial and business centres have become more and more important sources of PRC's FDI inflows. As shown in Table 4 [ PDF 64.7KB | 1 page ], their share of PRC's total FDI increased from only 0.3% in 1994 to 9% in 1998 and fell to 7.9% in 2001. For the period 1994-2001, the weighted average share of FDI by the offshore centre is as high as 9.6%. A significant part of FDI from the offshore centres could be round tripping FDI when the Chinese enterprises are attempting to use these centres to facilitate their financial transactions. But it is difficult to estimate directly the amount of round tripping FDI from the offshore centres. An indirect way to gauge this is to look at how other economies have used the offshore centres in facilitating their round tripping FDI. We are fortunate to have a clear direct estimation of round tripping FDI to Hong Kong, China from the offshore centres. The Hong Kong, China Government obtained these numbers from a detailed survey specifically designed to find out the extent of round tripping capital movement through the offshore centre. The results are not only relevant for Hong Kong, China but also can be illustrative for PRC as the offshore centres are primarily used for managing capital flows of listed companies traded in Hong Kong, China’s stock markets. There are no reasons why the Mainland companies, if they can move capital to these offshore centres in the first place, cannot move capital back to the Mainland as easily as Hong Kong, China companies do in their case of round tripping FDI. This is so because PRC does not have much restriction on FDI inflows in the form of FDI. Hence, it is reasonable to assume that on average the ratio of round tripping FDI from the offshore financial centres for the case of PRC could be similar to the ratio for the case of Hong Kong, China. Hence, the ratio of round tripping FDI through offshore centres in the case of Hong Kong, China provides a useful indicator for us to estimate the likely range of the similar ratio in the case of PRC.

Table 14 [ PDF 62KB | 1 page ] shows the estimation of round tripping FDI to Hong Kong, China through the offshore centres by the Hong Kong, China government statistics department. The ratio of round tripping FDI from offshore centres to Hong Kong, China was 40.4% in 1998, 27% in 1999, 48.3% in 2000, 14.4% in 2001, and 82.6% in 2002. The weighted average of this ratio for the period 1998-2002 is 40.1% and their standard deviation is 25.9%. This tells us that the round tripping FDI through offshore financial centres could be very large.

In the next sub-section, we will not use this ratio for estimating directly PRC’s round tripping FDI through offshore financial centres. Instead, we will use the case of Hong Kong, China to argue that PRC’s ratio of round tripping FDI through the offshore centres should be larger than the lowest ratio of round tripping FDI we estimated for PRC’s six FDI source regions. More specifically, we argue that for the rest of FDI source regions excluding U.S., Germany, Japan, Republic of Korea, Taipei,China Province, Singapore, and Hong Kong, China, the ratio of round tripping FDI to PRC is similar to that for Germany, e.g. around 22.4% or within a range of 13.8% to 31%.

The Scale of PRC's Round Tripping FDI

In the previous subsections we estimated directly from the available statistics the round tripping FDI to PRC from seven FDI source regions: U.S., Germany, Japan, Republic of Korea, Taipei,China Province, Singapore, and Hong Kong, China. Table 15 puts all the crucial information together in an attempt to estimate an average ratio of round tripping FDI in PRC in recent years. As shown in Table 15, for the year in 2000, according to PRC’s FDI statistics, the above seven regions contributed US$29.7 billion FDI to PRC, which is 72.9% of PRC’s total FDI of US$40.7 billion. Table 15 [ PDF 63KB | 1 page ] also provide the weighted average of the round tripping FDI ratio for the seven regions as a whole: 46.5% with a range from 34.9% to 58.1%.

Now, the problem is we don't have any direct information about the ratio of round tripping FDI for the rest of PRC's FDI source regions. An overly conservative approach to deal with this is to assume that there is zero round tripping FDI to PRC from the rest. If this assumption is used, then from simple calculation the weighted average of the round tripping FDI ratio for PRC as a whole would be 33.9% with a range from 25.5% to 42.4%.

A more reasonable approach is to assume that the round tripping FDI ratio for the rest regions to be the same as for the Germany since Germany has the lowest ratio. This assumption is likely to be conservative since regions like offshore financial centres are likely to have much high round tripping FDI ratio. When this assumption is used, PRC's overall round tripping FDI ratio is 40% with a range from 29.2 to 50.2% as shown in Table 15 [ PDF 63KB | 1 page ]. We believe this is the best estimate based on all the available information.

Our estimation shows clearly the scale of round tripping FDI in PRC is very large although the middle estimation of 40% for PRC's round tripping FDI is only one half of the 80% for the share of M&A related FDI for the global FDI flows. Table 16 [ PDF 63.9KB | 1 page ] compares our estimation of the general pattern of PRC's round tripping FDI with the pattern of PRC's capital flight as estimated by Gunter 2004. In Table 16 [ PDF 63.9KB | 1 page ], we multiply the high, middle and low estimates of the average ratio of PRC's round tripping FDI to PRC's total FDI as reported by PRC to get predicted flows of PRC's round tripping FDI during 1994-2001 for the high, middle and low estimates. The predicted flows of PRC's round tripping FDI are then divided by PRC's capital flight during the same period for high, middle, and low estimates respectively. The weighted average of the ratio of round tripping FDI over capital flight is 21.2% for the high estimates, 23.9% for middle estimates, and 30.6% for low estimates. In another word, based on the data during the period 1994-2001 about 20% to 30% of PRC's flight capital has returned back to PRC in the form of round tripping FDI. This seems a reasonable result to us.

In this paper we have tried to focus on finding out the overall scale of PRC's round tripping FDI since that is the most relevant information for policy debates. It would be useful to know how the round tripping FDI flows are affected by many specific factors over time such as changes in tax rates, expectations on changes in exchange rates, relaxation of capital control, access to overseas capital markets, rate of returns of investing in PRC etc. But the limited amount of the data does not allow us to investigate these interesting issues in any detail. By looking at the available data it seems reasonable to conclude that the major driver for the round tripping FDI is the long-term dilemma that on the one hand there are profitable opportunities in PRC but one the other hand investors would like to keep their capital aboard. Unlike the short term flows of portfolio capital or other speculative investment, FDI in PRC is relatively stable against the fluctuations in many of the macro economic variables such as interest rates, exchange rates, and tax rates. The relationship between PRC’s round tripping FDI and PRC’s capital fight seems also quite stable over the longer run. The scale of PRC’s round tripping FDI is large but not far from international experiences such as in the case of U.S. or in the case of cross-border M&A. Although the margin of errors for our estimation is large due to the inaccurate nature of FDI statistics, qualitatively there is no doubt that PRC's round tripping FDI is very large and significant since the data from PRC’s seven FDI source regions show the same consistent patterns and they together accounted for more than 70% of PRC's total FDI inflows.

Download this Discussion Paper [ PDF 535.5KB| 48 pages ].

Comment(s)

There are [1] comment(s) for this entry. Post a comment.

1. kundan singh
   (posted 04 September 2006 / 10:10:07 AM)
   
   
   
   Thank you for this paper on round tripping. I am currently writing my masters thesis on the reasons for growth of fdi in China and India. Round tripping plays a part in the limitations of this research. The information has been useful in my thesis.
   
   Thank you
   
   Kundan Singh