1
Capital control implementation in China
compared to the United Kingdom
Abstract
In this thesis the impact of capital control on the covered interest parity (CIP) for China and the United Kingdom is evaluated. China is a country with strict capital controls which is contrary to the United Kingdom which has an open economy without restrictions. The United States is used as a benchmark to assess the deviation of CIP. As in previous literature studies was confirmed, conducted before 2008 by Ma & McCauley (2007), capital controls in China remain binding during a period from 2008 until 2016. The United Kingdom showed that in absence of capital controls, CIP will hold.
Student: Mark ter Avest Student number: 10446206 Supervisor: Egle Jakucionyte Date of submission: June 2016
University of Amsterdam Faculty Economics and Business
2
Statement of Originality
This document is written by student Mark ter Avest, who declares to take full responsibility for the content of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the content.
3
Content
Introduction ... 4
Literature Review ... 7
Capital control ... 7
Onshore-offshore renminbi market ... 9
Methodology ... 12 Results ... 15 Conclusion ... 17 Discussion ... 19 References ... 20 Appendix ... 22
4
Introduction
China’s financial system is still very closed relative to other economies. In the article (He, Lillian, Wenlang, & Tommy , 2012) a comparison is made across 25 advanced and emerging economies with different degrees of capital account liberalisation. In table 1 of the appendix the economies are grouped into 4 classes according to the degree of capital account liberalisation, based on a period between 2004 and 2009. Class A are the economies with the lowest degree of capital account liberalisation, followed by classes B and C. Class D shows the economies with the highest degree of account liberalisations. China is ranked class A while the United Kingdom is ranked class D (He, Lillian, Wenlang, & Tommy , 2012). Table 2 shows that capital account openness will lead to an evident rise in international investment positions. China is the category with the least liberated economy and the least international investment positions measured relative to GDP while the United Kingdom is in the category with the most liberated economy and has the highest international investment positions. Although China is ranked class A, there are increasing signs the authorities are in favour of relaxing capital controls and promoting greater use of the Chinese currency abroad. Timescales are still uncertain, although full liberalization could potentially occur within a decade (Hooley, 2013).
China’s economic performance during the last quarter of the century has been one of the strongest in history. By the end of 2012, it has become the world’s second largest economy, a key global exporter and an important trading partner for developing and developed economies. China has also been a large receiver of foreign investment and it emerged as a leading investor in global financial markets. Alongside the obvious domestic and global benefits, this fast development has brought a number of policy challenges, and structural imbalances have emerged both internally and externally. Examples of these imbalances are overcapacity in industries, external surpluses, and risks of abundant liquidity. The imbalances are linked to the tightly managed exchange rate regime, which obstructs the performance of monetary policy (Cappiello & Ferrucci, 2008). To improve the imbalances China has implemented policy challenges, which also affected the capital account liberalization. Like many other developing economies, China confronts a complex set of questions related to the desirability and mode of implementing capital account liberalization as well as to domestic financial reform. In the past 15 years, incremental progress has been
5
made in liberalization of the capital account (Zhang & Gou, 2015). Under capital account restrictions, the People’s Bank of China (PBoC), China’s central bank, intervenes by purchasing foreign-currency or domestic bonds. Given this policy stance, the PBoC holdings on reserves have grown rapidly and China’s monetary policy has become increasingly sensitive to global financial conditions. (Chang, Liu, & Spiegel, 2015).
Unlike China, the United Kingdom has an open economy without regulatory capital control. Since the end of 1979 all the exchange controls were abolished (Roy, Misra, & Misra, 2006). It does not have tight restrictions placed on the flow of funds across its borders. In this paper the effectiveness of capital control will be evaluated. The comparison will be made between China and the United Kingdom. The United Kingdom is chosen because it is a class D country with an open capital account which is the opposite of China, which has a relatively closed economy, class A. A possible way to assess the effectiveness of capital controls in practise is by evaluating if there are deviations from the covered interest parity, CIP. To illustrate, this paper assesses whether the difference between the forward rate and the spot rate of two currencies is equal to the difference between the domestic interest rate and the foreign interest rate. A deviation would imply that there are arbitrage opportunities, unless there are capital controls preventing this. Previous literature from Cappiello & Ferruci (2008) and Shu, Su, Chang & Miao (2008) find that there are significant deviations from the CIP between 1999 and 2007. This proves that capital controls have restricted portfolio flows and have been binding. This paper will extend the calculations performed by Shu, Su, Chang & Miao (2008) by adding the United Kingdom to the model and look at the impact of capital control of both countries from 2008 until 2016. In the article mentioned above only China was evaluated. Interesting is, that in this paper the United Kingdom will be used as a country with a different degree of capital control compared to China. China, a country with capital control, alongside United Kingdom, a country with an open economy. In addition to analyse the capital control policy, the cross-border arbitrage opportunities in China will be measured. Such a test is based on onshore and offshore prices. The offshore renminbi referred to as CNH, does not fluctuate within a tight band like the onshore renminbi, CNY, and is free of the control of the PBoC. The offshore renminbi came about as China began trying to internationalize its currency. A variation of prices can indicate effectiveness of capital controls and the restrictions on arbitrage. This paper will provide an insight of to the
6
following question; what is the impact of capital control, measuring the deviation on CIP for China and the United Kingdom during a period between 2008 until 2016?
7
Literature Review
Capital control
Theoretical literature suggests that there are several benefits of capital account liberalisation. At first, liberalization improves capital allocation efficiency and risk diversification, and fosters financial development, which will stimulate economic growth (Kose, Prasad, Rogoff, & Wei, 2009). Secondly, it saves the transaction cost of capital controls.
There are also several arguments in favour of capital controls. First, large capital flows help to synchronise the domestic and international business cycles. Secondly, flows of capital can produce extra fluctuations in liquidity and make monetary policies ineffective (Levy, Schmukler, & van Horen, 2009) (Rodrik, 1998).
In July 2005 the PBoC announced that it wants to end the dollar peg and shift to a managed market-based floating rate regime. A flexible exchange rate regime is introduced to allow the currency to move within a narrow band around a central parity rate that is determined by a basket of world currencies. This process was disrupted in mid-2008 because of the global financial crisis. In June 2010 the PBoC announced that it was going to increase its flexibility and proceed with the reform of the RMB exchange rate regime. According to Eichengreen (2011), this market orientation of the Chinese exchange rate regime increases the likelihood that the RMB will become an international currency. According to the IMF, capital controls provides a basis for a “de jure” measure, which is more closely related to the intentions of a policy on its capital account. Opposed to this there are “de facto” measures which are associated with what actually happened. In China there a discrepancy between “de facto” and “de jure”, according to the annual IM report on Exchange Arrangements and Exchange restrictions. Different studies about China have pointed out that it does not always conform to the public announcement, which means there are considerable differences between policy announcement and its implementation (Dixon, Zhang, & Dai, 2016).
Besides the increase of liberalisation of the capital control in China, the flexibility of the exchange rate also increased over the past decade. According to Funke et al. (2015), the PBoC took an important step in 2005 when it implemented a managed floating exchange rate regime instead of a fixed regime. The new exchange rate is determined in reference to a basket of currencies instead of the US dollar alone. Over time this trading band has been
8
widened from 0.3% in 2005 to 0.5% in 2007. In 2010 the band increased to 1% and in march 2014 to 2%.
To control the capital market China uses a wide variety of instruments to set their own monetary policy. These instruments include setting administered deposit and minimum lending rates, quantitative measures like reserve requirements, lending quotas and administrative restrictions for investments. This thesis will look if it is possible to see if there is a direct impact on the CIP when new measures are implemented.
The covered interest parity theorem states that the covered interest differential between two identical assets denominated in different currencies should be zero. Deviations of the covered differential from zero represent riskless arbitrage opportunities and therefore indicate inefficiency in international capital markets because these profit opportunities go unexploited (Taylor, 1987). In order for interest parities to hold, several assumptions are required. First of all, there should be free capital mobility. Secondly, there should be an absence of transaction cost, which means there is no market hindrance to arbitrage across countries. Thirdly, there has to be no default risk, financial investment has to be safe against country risks. Even if CIP holds there might appear statistical discrepancy from various reasons. Time difference between various countries in different time zones or the certificate of deposits is not equal due to different liquidities.
The United Kingdom experienced rapid liberalization of capital controls in 1979. Two currency crises, which occurred in 1967 and in 1976, led to poor economic performance. This development in combination with a profitable reserve of foreign currency reserves led to the decision of the government for the abolition of capital controls. The government recognized that the abolition of capital controls had to be accompanied by domestic deregulation of macro-economic policies. In the end of 1979 all the exchange controls were abolished (Roy, Misra, & Misra, 2006). Even until today the United Kingdom has no restrictions on capital transactions in money, capital and derivatives market and with respect to personal capital transactions and institutional investors.
As a benchmark to calculate the CIP, one of the most developed stock markets is chosen to make a comparison between China and the United Kingdom, namely the United States (Ross & Zervos, 1998). The United States is chosen in this paper because it has liberated capital controls and a developed stock market.
9
The United States, just like the United Kingdom, has no restrictions on capital transactions toady. Most of the controls were eliminated from 1974 onwards after the breakdown of the Bretton Woods system. Since then, the United States has followed a liberal capital regime with limited controls, mainly maintained due to security concerns (Bakker & Chapple, 2002).
Onshore-offshore renminbi market
China started promoting the international use of its currency in 2003 when it developed offshore markets for the renminbi. The development of the international role of renminbi is tied to Chinese capital account liberalization. The BIS Triennial Central held a survey in 2013 on the foreign market exchange and concluded that the renminbi is ranked ninth in most traded currency. It has a daily turnover of 119.6 billion US dollar. China’s role in the money market is becoming more important within Asia. The offshore renminbi was founded by creating the first offshore settlement infrastructure and personal renminbi banking services in Hong Kong. Multiple studies, eg Henning (2012) and Subramanian & Kessler (2015), find that, since China moved to a managed float regime in July 2005, the renminbi’s impact on the Asian currencies has become increasingly significant. The renminbi is traded onshore and offshore. Onshore, or the CNY currency, is subjected to the central bank’s intervention. CNH is traded offshore, which means it is traded outside China on a foreign market. This market is not subjected to the Chinese central bank’s intervention or its fluctuations resulting from the daily trading band of the rate movement. The exchange rate is determined by market forces, unrestricted by the intervention of the PBoC or the Hong Kong Monetary Authority.
One way to manage the internationalisation of the renminbi was by the introduction of QFII and QDII schemes. QFII stands for qualified foreign institutional investor scheme and was introduced in 2003. This marked the beginning of the offshore market of the RMB. Before this time it was not possible for foreign investors to invest in Chinese equity and bonds, the only exception was the foreign currency B-share market. QDII stands for Qualified domestic institutional investor which permits Chinese individuals and companies to invest in overseas securities. The QFII and QDII may interact to help the liberalisation process. Bigger portfolio inflows via QFII and a relaxation of the QDII outflows help to make capital controls less binding over time (Ma & McCauley, 2007). China’s capital
10
account was ranked among the least open in the world according to the Chin-Ito index, a commonly used indicator for capital account openness (Chinn & Ito, 2006). The process of RMB capital account convertibility was interrupted by the global financial crisis. This was due to shortage of hard currencies and increased volatility among major currencies (Prasad, 2016). After the global financial crisis in 2008, the PBoC introduced two reformed channels for the inflow and outflow of capital. For the inflow of capital, the Renminbi Qualified Foreign Institutional Investor (RQFII) scheme was launched in 2011. This allowed qualified institutions to use offshore RMB fund to invest in Chinese equity. Another channel for capital outflows was introduced, the qualified domestic individual investor (QDII2) scheme, which was proposed in 2013 but is not launched yet. This will permit individual retail investors to invest 1 million RMB in offshore financial products. Apart from the channels for inflow and outflow of capital mentioned above, China launched two-channel flows. The two-channel flows are divided into three categories. The first one, free trade zone’s (FTZs), was implemented in September 2013 and expanded in April 2015. Cross border capital transactions within these zones have been liberized. The second is the Shanghai-Hong Kong Stock Connect which was launched in 2014. The third flow is the mutual fund connect (MFC) which launched in July 2015. This allowed Hong Kong and mainland funds to be distributed in each other’s markets through a vetting process. More details are provided in table 3.
Deviations exist between the onshore and offshore which results from the effectiveness of capital controls and the restrictions on arbitrage. Two main sets of factors can cause the gap between CNY and CNH. The first are related to capital market liberalisation policies. China has introduced several incoming and outgoing flow channels described above. These channels are introduced to foster the development of both onshore and offshore renminbi markets. The second factor can be the general market and economic conditions. The CNY and CNH market might respond differently to changes due to other investor bases and liquidity conditions (Funke, Shu, Cheng, & Eraslan, 2015).
Ma & McCauley (2007) have calculated the spread between onshore and offshore NDF-implied yields from 2004 until 2007. Their finding is that when there was a policy change in 2005 there was a direct change in the spread of the onshore and offshore market. The authors point out that the wide differences between onshore and offshore renminbi interest rates point
11
to the efficacy of capital controls. The experience with the onshore forwards contracts in 2006 pointed out that maintaining effective capital controls gets harder with the development of financial markets and further deregulation of cross-border transactions. This thesis will look at the difference between onshore and offshore renminbi rates between 2011 and 2016.
12
Methodology
There are different ways of assessing the effectiveness of capital controls. This paper will conduct multiple tests to examine the impact of capital control for China and the United Kingdom.
The first test is to measure deviations for the covered interest parity (CIP). In order to have an efficient market, free capital movement is an important condition. The deviation will be measured for China and for the United Kingdom separately. In an open and efficient market, CIP holds, so that an investor is indifferent between placing money on domestic account or foreign account. Naturally, CIP holds in economies with developed financial markets and open capital accounts. The investor is indifferent between investing in a foreign asset and earning the foreign interest rate, or converting the money into the local currency at the spot exchange rate and earning the same interest rate. The formula to measure CIP is:
𝑓"
𝑠" = (1 + 𝑖")/(1 + 𝑖"∗) (1) In the first situation the deviation of China is measured. China is used as home country and the United States as foreign country. In the second situation the CIP of the United Kingdom is measured where in this case the United States will also be chosen as foreign country and the United Kingdom as the home country. Hence, the United states is used as benchmark country in this test. The formula which measures the capital control premium is derived from the CIP as the deviation of the forward premium:
𝜉" = 𝑓"− 𝑠" − 𝑖"− 𝑖"∗ (2)
In the first situation where the capital control premium of China is calculated 𝑓" is logarithm
of the forward rate of the one period-ahead-rate and 𝑠" is the logarithm of the spot exchange rate at time t, expressed in terms of units of the domestic currency. In this case, the Renminbi is measured in terms of Dollars, which is from the foreign country, the United States. Just like the calculations performed by Shu, Su, Chang & Miao (2008) the onshore market is used for China because this market is more applicable to the domestic financial sector in general. The time span t for the CIP is set equal to one year. For 𝑖" the 12-month onshore interest rate swap, IRS is taken as the domestic interest rate. IRS is used because it is actively traded, whilst interbank lending concentrates at the daily and weekly maturities and with few
13
transactions at the one-year horizon. For 𝑖"∗ , the foreign interest rate, the 12-month United
States LIBOR is used. The measures are performed on daily basis ranging from December 2008 until May 2016.
In the second situation the deviation of the capital control premium is measured for the United Kingdom. Both forward rate (𝑓") and spot rate (𝑠") are expressed in terms of units
of the domestic currency, in this case Pounds, per unit of foreign currency, Dollars. 𝑖" is the domestic interest rate. In this situation of the United Kingdom the 12-month GBP LIBOR is used as interest rate. For 𝑖"∗ , the foreign interest rate, the 12-month United States LIBOR is
used. A deviation would imply that there are arbitrage opportunities, unless there are capital controls preventing these.
In the second test the following regression is estimated for both situations:
𝑓"− 𝑠"= 𝛼 + 𝛽 𝑖"− 𝑖"∗ + 𝜀 (3)
Where 𝑓" and 𝑠" are the logarithm of the forward and spot rates for each country compared to the United States. The part (𝑓"− 𝑠") is commonly referred to as the forward margin, while (𝑖"− 𝑖"∗) is often called the interest rate differential. For the foward 𝑓" 12-month exchange
rate is used measured in terms of units renminbi/GBP per US dollar. The interest rate differential consists of 𝑖", the 12-month domestic Chinese Onshore IRS/GBP LIBOR, and 𝑖"∗,
the 12-month LIBOR.
To test whether or not the CIP holds, the estimated coefficient for 𝑖"− 𝑖"∗ will be tested if
it’s equalling 1 by a Wald test, which is defined as follows: (𝜃 − 𝜃5)6
𝑣𝑎𝑟(𝜃) (4)
Were 𝜃5 will be defined as 1 and 𝜃 will be the estimated beta from regression (4). This test
follows a Chi-squared distribution. Several tests will be conducted. Different time series will be tested. The first test is based on the period from June 2008 until May 2016. In addition, an extra test can be performed to measure shorter periods where CIP might hold. The time range will be decided after the results of the test on deviation of the forward premium from CIP calculated earlier with formula (2).
14
The third test is to see if there is a wedge between the onshore and offshore renminbi market. This is an additional test to evaluate the effectiveness of capital control. This test can only be performed on China because the United Kingdom only has an onshore market, because there is no capital control by the Bank of England. The null hypothesis is that there are no substantial differences between renminbi interest rates onshore compared to offshore non-deliverable forward (NDF) exchange rates with the US dollar on the other hand. Large yield gaps suggest significant cross-border market segmentation and thus binding capital controls (Ma & McCauley, 2007). Since 2004 a daily auction of PBoC bills makes it possible to compare domestic renminbi money market yields with the renminbi yields implied by the NDF’s traded offshore. Daily data is used to calculate the onshore/offshore yield:
𝑆ℎ𝑜𝑟𝑒 𝑦𝑖𝑒𝑙𝑑 = 𝑓𝐶𝑁𝐻"− 𝑓𝐶𝑁𝑌" (5) The 𝑓𝐶𝑁𝐻 is the forward offshore exchange rate measured in renminbi per US dollar, which is traded on a forex market outside of the US. This paper uses the Hong Kong forex exchange rate. The 𝑓𝐶𝑁𝑌 is the forward onshore exchange rate measured in renminbi per US dollar. The forward rate is calculated on a 12-month forward period. The dataset points are averaged weekly. The first data point is August 2011 and the last data point is the May 2016. To test if the shore yield is significantly different from zero, a two-sided t-test will be used:
𝑇 =HIJK
L/ M (6)
The null hypothesis is that the shore yield will be equal to zero. Χ is the sample mean, S is the sample standard deviation of the sample and n is the sample size. 𝜇5 is set equal to zero in this case.
15
Results
In the first test the deviation of the forward premium of CIP is calculated for the United Kingdom and for China. In graph 1 China is displayed and in graph 2 the United Kingdom. The capital control premium is defined as the deviation of the forward premium from CIP. First looking at the graph from China, it is volatile and deviates between -0.02 and 0.013 percent. Between the periods, July and September 2007, June 2008 and December 2008, November 2014 and January 2015, March until May 2016 the capital control premium is close to zero. Looking at graph 2 the deviation from CIP of the United Kingdom is displayed. The CIP is around zero during the whole period and fluctuates between -0.002 and 0.003 percent. A positive control premium indicates that the covered returns on foreign assets are higher than the returns on domestic assets. This can be rearranged into:
𝜉"= 𝑓"+ 𝑖"∗ > 𝑠
"+ 𝑖" (7)
This means that capital controls prevent domestic capital from flowing out of the country. This situation has characterised China between a period from September 2008 until June 2009 and from October 2014 until today. Conversely, in a period from June 2009 until October 2014 the capital controls have prevented foreign capital from flowing into the country. Comparing this to the situation of the United Kingdom the volatility is between -0.002 and 0.003 percent. The range in China is 0.033 and in the United Kingdom it is 0.005 percent. This means the deviation of CIP is 6 times higher in China compared to the United Kingdom. From a time series point of view, the yield gap shows clear evidence of volatility clustering; there were extended periods when its volatility was high and there were periods of low volatility.
The results of the second test where the regressions are performed on China are shown in Table 5. The test is performed for several periods. In the first period the overall timespan is used which is from June 2008 until May 2016. In equation (3), if 𝛼 = 0 and 𝛽 = 1 in the estimated equation, the CIP is said to hold. The results show that CIP holds at the time period from Nov 14 until Jan 15 at a 5% significance level. For the rest of the tested periods, the hypothesis that CIP equals 0 can be rejected. Hence, CIP does not hold for China in most of the tested periods.
16
In addition to China the Wald test is conducted for the United Kingdom. The results are shown in table 6. The time period evaluated marks from June 2008 until May 2016. In the regression 𝛼 is insignificant at the 5% level. But the Wald test shows that 𝛽 equals one. At a 5% significant level we can conclude that CIP holds for the United Kingdom.
The third test performed in this paper measures the yield gap between CNY and CNH, performing a double sided t-test. The results are shown in table 7. There were 253 observations between a period from August 2011 until June 2016. The results are plot in graph 4. From August 2011 until August 2012 the one year PBC bill yielded on average 450 basis points more than the yield implied by the offshore NDF. After this period the yield decreased and eventually becomes negative in May 2013. After February 2014 there is an increase again of the yield which peaks at the first 3 months of 2015. At the highest point there is a yield difference of 1200 basis points. At the end of August 2015 the yield decreases to around zero and eventually becomes negative. The pricing gap is volatile. With the t-test the 𝜇5 is set equal to zero to test if CIP holds. The result shows that the mean is 3.29 which is far above zero. With a P value of 0.0000 and a significance level of 5% the null hypothesis is rejected. The results are insignificant to conclude that CIP holds. This indicates that the offshore renmimbi is an imperfect substitute for the onshore renmimbi. The differential is often large which makes firms and investors averse of using CNH as a substitute for CNY. This will hold back the development of an international role of for the renmimbi.
In addition to the test measuring the whole period from August 2011 until May 2016 a second tests is performed on the period from July 2015 until May 2016. This test is to see if the CIP holds after the implementation of Mutual Fund Connect. The graph shows that the yield difference decreased from 1200 basis points to almost zero. The result as shown in table 8 reveals that the mean is -1.15 and the P value is 0.0005. At the 5% significance level the null hypothesis is also rejected for the second test. Although it is not significant that CIP holds, the yield gap between the offshore and onshore renminbi decreased in the last period. The p-value increases from 0.0000 to 0.0005. This shows that the introduction of MFC definitely had an impact on the CIP.
17
Conclusion
Looking at the results it can be concluded that capital controls in China remain binding. The PBoC imposes administrative controls on the capital account. This is done to sterilise its economy from foreign shocks. CIP holds in the United Kingdom but in China there is a significant deviation from the CIP. This deviation in CIP cannot be held fully accountable by capital controls. Interesting to see from the results is that CIP holds in the United Kingdom, a country with an open capital account and without regulations or capital control instruments. China is slowly opening its economy in which free financial flow of capital has been an important element. The PBoC is trying to maintain a gradual approach freeing up the exchange rate, which creates large and volatile movements of capital which we have seen in the deviation of CIP and in the onshore offshore yield differential.
The offshore renminbi market has created an extra opportunity to evaluate the capital account control policy. The outcome of the second test is in line with the results of the first test. The second test shows that the onshore rates differ significantly from their onshore counterparts. This confirms that the capital controls are still binding. As discussed earlier, two factors for the volatile yield gap. The first factor is related to capital market liberalisation policies. The second factor can be the general market and economic conditions. Since the offshore rate differs significantly from its onshore counterpart this can reflect capital controls. Further it can also show differences in the investor composition in different markets and also contrasting sensitivities of investors to the same shocks. However, until now, little research has been done in identifying the difference in the Onshore Offshore yields.
Since the start of the measuring period used in this paper, June 2008, the CIP of China fluctuated heavily. Evaluating the CIP of China it can be concluded that capital is not the only determinant for the CIP parity. Multiple other instruments that are used by the PBoC to protect the Chinese economy from foreign shocks can be of influence that obstructs the CIP to hold. From these measurements can be concluded that if there is no capital control the CIP holds, but not vice versa.
China has implemented several channels to liberalize cross-border capital flows. The most important channels which have been implemented after 2008, RQFII and QDII2, have not shown a direct impact on the deviation of CIP. Although the implementation of the
18
Mutual Fund Connect decreased the yield gap from 1200 basis points to almost zero. This had a direct influence on the onshore offshore differential, but it is insignificant to conclude that CIP holds after the introduction of MFC. The impact of the different channels implemented by the PBoC are not sufficient for the CIP to hold or close the onshore offshore yield gap. This can be explained because of the difference between “de jure” and “de facto”. China’s implementations of the policies on its capital account do not show what actually happened. This discrepancy between “de facto” and “de jure” makes is difficult to assess which policies affected the CIP of China.
19
Discussion
To check the calculations performed for the first test in this article the results are compared to the results in Chart 2. from (Shu, Su, Chang, & Miao, 2008). In this article the period evaluated was from February 2006 until October 2008. After demeaning each variable in the equation in formula (2) the results from the graph are similar but the range is different. This is the result of a different normalisation. The results showed the same trend in a similar period of time, which can be seen in graph 3.
The regression and Wald test showed that CIP did not hold in China for a period from June 2008 until May 2016. This suggests that capital controls have been effective for the whole period. In the United Kingdom CIP holds since June 2008. This means that CIP is influenced by capital controls. Another way to interpret the results for the deviation in CIP is that other factors are of importance of the CIP differential. In addition to capital control the fixed exchange rate regime might also be of influence on the CIP. Further research will be necessary to indicate which factors might be of influence on CIP in addition to the capital control restrictions implemented by the PBoC.
This paper is limited by the information of the capital control channels of China. The limited openness of the PBoC makes it difficult to evaluate when capital control measures are exactly implemented. More detailed information of the announcement dates and implementation dates of capital market liberalisation policies of the PBoC would improve the calculations to examine the factors which cause the volatile yield gaps.
20
References
Bakker, A., & Chappel, B. (2002). Advanced Country Experiences with Capital Account Liberalisation. IMF Occasional Paper(214).
Calvo, G. A., & Reinhart, C. M. (2002). Fear of Floating. The Quarterly Journal of
Economics, pp. 379-408.
Cappiello, L., & Ferrucci, G. (2008, March). The sutainability of China's exchange rate
policy and capital account liberalisation. Retrieved from European Central Bank.
Chang, C., Liu, Z., & Spiegel, M. M. (2015). Capital controls and optimal Chinese monetary policy. Journal of Monetary Economics, 74, 1-15.
Chinn, M., & Ito, H. (2006). What matters for financial development? Capital controls, institutions, and interactions. Journal of Development Economics, 81(1), 163-192. Dixon, R., Zhang, Z., & Dai, Y. (2016). Exchange Rate Flexibility in China: Measurement,
Regime Shifts and Driving Forces of Change. Review of International Economics, pp. 4-10.
Eichengreen, B. (2011). The Renminbi as an International Currency. Journal of Policy
Modelling, pp. 723-730.
Funke, M., Shu, C., Cheng, X., & Eraslan, S. (2015, February). Assessing the CNH-CNY pricing differential: role of fundamentals, contagion and policy. BIS working papers. Hatzvi, E., Meredith, J., & Nixon, W. (2015, December). Chinese Capital Flows and Capital
Account Liberalisation. Bulletin, 39-48.
He, D., Lillian, C., Wenlang, Z., & Tommy , W. (2012, April). How would capital account liberalization affect China's capital flows and the renminbi real exchange rates?
HKIMR Working Paper, 09.
Henning, C. (2012). Choice and coercion in East Asian exchange rate regimes. Working
paper, 12-15.
Hooley, J. (2013). Bringing down't the great wall? Global implications of capital account liberalisation in Chian. Quarterly Bulletin, 4, 1-12.
21
Kose, M. A., Prasad, R., Rogoff, K., & Wei, S. J. (2009). Financial globalization: A reappraisal. IMF Staff Papers 561, 8-62.
Levy, y., Schmukler, S. L., & van Horen, N. (2009). International financial integration through the law of one price: The role of liquidity and capital controls. Journal of
Financial Intermediation, 183, pp. 432-463.
Ma, G., & McCauley, R. N. (2007, August). Do China's capital controls still bind? Implications for monetary autonomy and capital liberalisation. BIS Working
paper(233).
Prasad, E. S. (2016). China's Economy and financial Markets: Reforms and Risks. Hearing
on "China's 13th Five-Year Plan" (pp. 1-15). Brookings Institution.
Rodrik, D. (1998). Who needs capital-account convertibility? Essays in International
FInance(207).
Ross, L., & Zervos, S. (1998). Capital control liberalization and stock market development.
World Development(26.7), pp. 1169-1183.
Roy, M., Misra, R., & Misra, S. (2006). A Review of Cross-Country Experience in Capital Account Liberalisation. Reserve Bank of India Occasional Papers, 27(1), 2-10.
Shu, C., Su, X., Chang, J., & Miao, H. (2008, December). China's Monetary Regime in dealing with large BoP surpluses: framwork, effectiveness and costs. China Economic
Issues, 1-50.
Song, Z., Storesletten, K., & Zilibotti, F. (2014). Growing (with capital controls) like China.
IMF Economic Review, 62(3), 327-370.
Subramanian, A., & Kessler, M. (2015). The renminbi bloc is here: Asia down, rest of the world to go? Working paper, 12-19.
Taylor, M. P. (1987, November). Covered Interest Parity: A High-Frequency, High-Quality Data Study. Economica, 429-438.
Zhang, L., & Gou, Q. (2015). Capital Account Liberalisation in China. In L. Song , & R. Garnaut, China's domestic transformation (p. 239).
22
Table 1. Capital Account Openness across Economies
Class A Class B Class C Class D
China Australia Belgium Hong Kong SAR
India Brazil Chile Italy
Indonesia Finland France Netherlands
Malaysia Russia Germany Peru
Mexico Turkey Japan United Kingdom
Thailand Venezuela Korea Singapore United States
Sources: IMF's Annual Report on Exchange Arrangements and Exchange Restrictions 1997-2010 and HKMA staff estimates
Table 2. Capital Account Openness and International Investment Positions across Economies
% of GDP FDI assets FDI Libilities Portfolia Assets Portfolio Liabilities (Standard deviations in brackets)
Class A 5.20% (7.2%) 20.70% (15.3%) 3.00% (3.4%) 14.90% (11.1%)
Class B 15.80% (12.9%) 20.00% (6.4%) 25.30% (22.0%) 36.80% (31.9%)
Class C 35.10% (32.4%) 43.60% (40.6%) 58.50% (45.8%) 49.10% (30.1%) Class D 36.30% (28.3%) 32.60% (22.0%) 51.80% (33.1%) 54.20% (34.3%) Notes: The avesrage reatios of international investment positions to GDP are illustrated for each class of economies with the same captial account opennes. Standard deviations are also in % of GDP
Sources: IMF, IFS, CEIC and HKMA staff estimates
Appendix
23
Table 3. Summary of recent schemes to liberalize cross-border capital flows
Channels for Inflows
Qualified Foreign Institutional Investor (QFII) Scheme: Launched in 2002. Allows
qualified foreign institutions to convert foreign currency into RMB and invest in Chinese equities (both A shares and B shares) and a range of other RMB denominated financial instruments. As of October 2015, a total quota of $78.9
billion had been granted to 277 foreign institutions, including 8 central banks and 10 sovereign wealth funds.
Renminbi Qualified Foreign Institutional Investor (RQFII) Scheme: Launched in
2011. Allows qualified institutions to use offshore RMB funds to invest in Chinese equities and other RMB-denominated financial instruments. As of July 2015, a total quota of $68.4 billion had been granted to 135 financial institutions.
Channels for Outflows
Qualified Domestic Institutional Investor (QDII) Scheme: Launched in 2006.
Allows Chinese domestic financial institutions—commercial banks, securities companies, fund management companies, and insurance companies—to invest in offshore financial products such as securities and bonds. As of November 2015, a total quota of $90 billion had been granted to 132 financial institutions.
Qualified Domestic Individual Investor (QDII2) Scheme: Proposed in 2013; not
yet launched. Will permit individual retail investors with at least RMB 1 million ($160,000) in assets to invest in certain offshore financial products.
Channels for Two-Way Flows
Free Trade Zones (FTZs): Shanghai FTZ launched in September 2013. Three
new FTZs in Guangdong, Tianjin, and Fujian launched in April 2015. The FTZs use a “negative list” approach to regulate foreign investment—there are few restrictions on foreign investment in industries not on the list. Cross-border capital transactions and establishment of financial institutions within the zones have been liberalized.
Shanghai-Hong Kong Stock Connect: Launched in 2014. Allows mainland
Chinese investors to purchase shares of select Hong Kong and Chinese
companies listed in Hong Kong, and lets foreigners buy Chinese A shares listed in Shanghai. HK-to-Shanghai annual quota: RMB 300 billion ($47 billion); daily quota RMB 13 billion ($2 billion). Shanghai-to-HK annual quota: RMB 250 billion ($39 billion); daily quota: RMB 10.5 billion ($1.6 billion).
Mutual Fund Connect: Launched in July 2015. Allows eligible mainland and
24
streamlined vetting process. Initial aggregate investment quota: RMB 300 billion ($47 billion) each for inward and outward fund flows.
Source: Retrieved from (Prasad, 2016)
Table 4. IMF exchange arrangements and exchange restrictions
Effective March 1, 2011, the SAFE permitted qualified foreign-exchange-designated banks to launch renminbi (RMB)–foreign exchange cross-currency swap operations for customers. Easing China
Effective April 1, 2011, the SAFE permitted the interbank foreign exchange market to introduce trading of RMB versus foreign exchange options and banks to handle options services for customers. Easing China
Effective December 1, 2011, the SAFE permitted banks to handle RMB versus foreign exchange option portfolio operations. Easing China
Effective April 16, 2012, the floating band of the RMB’s trading prices against the U.S. dollar in the interbank foreign exchange market was widened from 0.5% to 1%—i.e., on each business day, the trading prices of the RMB against the U.S. dollar in the interbank foreign exchange market may fluctuate within a band of ±1% around the central parity released on the same day by the China Foreign Exchange Trading System. The spread between the RMB and the U.S. dollar selling and buying prices offered by foreign-exchange-designated banks to their customers may not exceed 2% of the central parity (previously 1%).
Effective March 17, 2014, the floating band of the renminbi’s (RMB’s) trading prices against the U.S. dollar in the interbank foreign exchange market was widened from 1% to 2%. That is, on each business day, the trading prices of the RMB against the U.S. dollar in the market may fluctuate within a band of ±2% around the central parity released that day by the China Foreign Exchange Trade System. The range of the spread between the highest offer price and the lowest bid price for RMB–U.S. dollar spot transactions at foreign-exchange-designated banks and their customers was widened from 2% to 3% of the central parity.
Effective July 14, 2014, the middle rate of the renminbi against the pound sterling is determined based on the average of the day’s market makers’ quotes. Previously, the rate was determined through the cross-rates by the China Foreign Exchange Trade System based on the day’s middle rate for the renminbi against the U.S. dollar and the exchange rates for the U.S. dollar against the pound sterling.
25
One-Sample t test
Variable Obs Mean Std. Err. Std. Dev. [95% Conf. Interval] CNH-CNY 253 3.298093 0.2434641 3.872534 2.81861 3.777577 t = 13.5465 252 0.0000 Table 8. T-test CNH CNY July 2015 until May 2016 One-Sample t test
Variable Obs Mean Std. Err. Std. Dev. [95% Conf. Interval] CNH-CNY 42 -1.153691 0.3061249 1.983916 -1.771922 -0.5354588 t = -3.7687 41 0.0005 Pr(|T| > |t|) = Table 7. T-test CNH CNY Pr(|T| > |t|) = degrees of freedom = degrees of freedom =
Sources: Reuters, Bloomberg and authors estimations Table 5. Test of CIP in China
(June 08 - May 16) (Juli 07 -Sept 07) (June 08 - Dec 08) (Nov 14 - Jan 15) (March 16 - May 16) Constant -0.000193 -0.03836 -0.0093336 -0.0000443 0.001912 -0.84 -5.84 -28.28 -0.05 2.4 0.3335184 -2.456444 -2.572374 1.039187 0.6049791 13.03 -3.21 -20.43 11.42 3.51 Adjusted R^2 0.0751 0.1746 0.7634 0.7551 0.2082 Number of observations 2080 45 130 43 44 Wald test for β=1 677.82 20.41 804.67 0.19 5.25 p-value 0.0000 0.0000 0.0000 0.6689 0.0271 Table 6. Test of CIP in United Kingdom (June 08 - May 16) Constant -0.0000372 -5.38 0.9953218 402.58 Adjusted R^2 0.9873 Number of observations 2080 Wald test for β=1 3.58 p-value 0.0586
26
Graph 1. Deviation of the forward premium from CIP of China
Sources: Reuters, Bloomberg and authors estimations
Graph 2. Deviation of the forward premium from CIP of the United Kingdom
Sources: Reuters, Bloomberg and authors estimations
-0.025 -0.015 -0.005 0.005 0.015 0.025 June
08 June 09 June 10 June 11 June 12 June 13 June 14 June 15
%
Deviation CIP China
Deviation CIP China -0.025 -0.015 -0.005 0.005 0.015 0.025 June08 June 09 June 10 June 11 June 12 June 13 June 14 June 15
%
Deviation CIP United Kingdom
Deviation CIP United Kingdom
27
Graph 3. Deviation of the forward premium from CIP of China similar to (Shu, Su, Chang, &
Miao, 2008) from August 2006 until October 2008
Sources: Reuters, Bloomberg and authors estimations
Graph 4. China’s onshore offshore yield gap
Sources: Reuters, Bloomberg and authors estimations
-0.05 -0.04 -0.03 -0.02 -0.01 0 0.01 Au g 06 Oc t 0 6 De c 06 Fe b 07 Ap r 0 7 Ju n 07 Au g 07 Oc t 0 7 De c 07 Fe b 08 Ap r 0 8 Ju n 08 Au g 08
Deviation CIP China
Deviation CIP China -10 -5 0 5 10 15August 11 August 12 August 13 August 14 August 15
CNH CNY yield 12 month
CNH CNY yield 12 month
