Research On Cross-border Capital Flow Management In China

During the Asian financial crisis in 1997 and the global financial crisis in 2008,the sharp fluctuations and directional reversals of cross-border capital flows severely impacted the financial stability of emerging countries,triggering currency crises and financial market turbulence.In the post-crisis era,cross-border capital flows have shown new characteristics of increasing fluctuation and shortening fluctuation cycle,making the financial stability of developing countries under constant pressure.Therefore,cross-border capital flow management is a problem faced by most emerging economies.In particular,sudden large-scale and continuous capital flows will pose challenges to a country through exchange rate,liquidity,credit and other aspects.As the largest emerging market in the world,China is in the process of expanding its opening-up and experiencing the cycle of capital inflow and outflow along with the cycle of international capital flows.Cross-border capital inflow and outflow gradually become an important factor that affects China’s financial stability.Based on the analysis of the international view evolution about capital flow management,policy practice and difficulties,this paper firstly studied the structural characteristics of procyclical of cross-border capital flows through country panel data to provide an empirical basis for the subsequent selection of risk variables and risk identification criteria.On this basis,combined with the application of big data integration machine learning and variable selection methods,this paper tries to explore and improve the risk early warning system,management philosophy and framework of cross-border capital flows.The main research results are as follows.Firstly,this paper analyzes the evolution of viewpoints,policy practices and difficulties in the management of international cross-border capital flows,and proposes to provide solutions by applying new technologies integrated with big data.After the international financial crisis and the impact of the "paradox of duality",IMF put forward a four-level framework for crossborder capital flow management in 2012,namely,structural policy,macroeconomic policy,macro-prudential policy and capital flow management measures,and put forward operating principles for different risks.Although these views have guiding significance,as cross-border capital flows themselves are non-discriminatory,certain contradictions and dilemmas are bound to exist in the management process of developing countries.On one hand,it is difficult to form an agreed management framework,one the other hand,there is the conflict between macroscopic and microscopic means in reality.Therefore,there is no one-size-fits-all framework for the selection of management policies.So it is necessary to select a policy toolbox suited to national conditions.Combined with the reality of China,China proposes a crossborder capital flow management framework of "macroprudential + microcosmic supervision".However,from the practice in recent years,China’s cross-border capital flow management is still inadequate: on the one hand,it is difficult to identify the short-term capital flow shock in advance,because predicting economic turning points in advance is difficult,so risk warning mechanisms often fail to play a good role.On the other hand,how to coordinate alternative policies for managing risk.In reality,it is difficult to separate macro-control policy and capital control.Second,depicting the risk of cross-border capital flow from three dimensions: exchange rate,liquidity and foreign exchange gap of banks.1.From the perspective of exchange rate,introducing the risk index of ex-ante market pressure,that is,the change of exchange rate in the case of rational expectations and the central bank not intervening.Ex-ante exchange market pressure are the leading indicators of the exchange rate changes,excluding the control factors of the monetary authorities,so the influencing factors and changes are also more dependent on the economic fundamentals and international monetary policies.2.From the perspective of liquidity,introducing money creation risk,this is because the disposal of foreign exchange liquidity is ultimately reflected in the corresponding money creation risk,that is,the central bank spit out base money,foreign exchange loans of commercial banks are reflected in money creation;3.This paper introduces the risk index of foreign exchange supply and demand gap to measure the currency mismatch risk of the banking sector.The study on the relationship between cross-border capital flows and the three types of risks shows that the three types of risks had a long-term co-variation relationship with cross-border capital flows,and the granger causality test showed that the three types of risks were significantly driven by cross-border capital flows,and were relatively independent,so it is necessary to conduct targeted analysis.Thirdly,this paper studies the characteristics and procyclicality of China’s cross-border capital flows.The risk of cross-border capital flow originates from the procyclicality.The research on the procyclicality is an important part of the overall logic of this paper.From the perspective of the characteristics of cross-border capital flow and related risks,during the period of cross-border capital outflow,all of the three types of induced risks have greater volatility;From the perspective of procyclicality test,it is found that the procyclicality of China’s crossborder capital flow mainly comes from the domestic economic cycle,while the response to the international financial cycle is not obvious;From the perspective of the three types of risks,during the period of cross-border capital outflow,the influence of internal economy on the three types of risks is obviously stronger than inflow period.Once the domestic economic growth slows down the risks related to outflow will become more prominent.Therefore,in the subsequent early warning of risk events,the risk pressure standard is set as the outflow related risk exceeds a certain threshold,so the risks are respectively set as the risk of depreciation pressure in the foreign exchange market,the risk of currency contraction and the risk of foreign exchange deficit.Fourth,Big data and machine learning algorithms can provide more timely and accurate prediction results in cross-border capital flow monitoring,helping to find the "anchor" of crossborder capital flow macro-prudential management.This paper derived more than 70000 variables to build big data monitoring database from China’s macro economy,financial markets,international financial etc.,and apply Adaboost method to predict the risk of cross-border capital flows in China,successfully realized the risk prediction of the three dimensions in the next month,the accuracy of the model is up to 70-80%.Among them,macroeconomic fundamentals and domestic capital market fluctuations are the most important characteristic indicators to predict China’s ex ante EMP.The interbank market rate is the most important leading indicator of the risk of monetary contraction.Foreign economic and trade situation is the most important leading indicator of FXSLP.Fifth,the implementation of relevant management policies after risk monitoring.Based on the reality of that cross-border capital flows caused by the different have different risk management object,orientation,policy tools.Based on the parameters of the three types of risk models,this paper determines the management orientation of the three types of risks,and studies the management policy combination of the three types of risks through adaptive Lasso model,taking the relevant policies of China as samples.Through a comprehensive comparison of the effects of various management policies,it is found that the effects of money multiplier and interest rate spread are gradually weakening,while macro-prudential policies such as central parity management,floating range management of exchange rate and foreign exchange credit management have good effects;In exchange rate policies,managing the central parity rate and expanding the floating range of exchange rate have good effect.Among the macroprudential measures,foreign exchange credit management has good effect,in particular,it is the key point in the extreme outflow situation.The results of this paper are of certain value to the establishment of a timely,accurate and effective risk management system for cross-border capital flows,which follows international conventions and is based on China’s reality:First of all,it is about the construction of the risk index: 1.the Using ex-ante EMP to replace the traditional EMP index as the measurement of exchange rate risk,which excludes the control factors of monetary authorities;2.Constructing the theoretical model of "foreign exchange flow-money creation" from the perspective of liquidity;3.From the perspective of the banking sector,introducing the index of the balance ratio of foreign exchange settlement and sale,which can reflect the risk of bank currency mismatch and the willingness of enterprises to sell foreign exchange.Second,Exploring the application of machine learning and variable selection methods combined with big data.In terms of risk prediction,the big data method adopted in this paper widely obtain data related to cross-border capital flows,which can select indicators with optimal prediction ability,and establish a characteristic index system for predicting the risk of cross-border capital flows.Third,this paper conducts a comprehensive comparison of the policy framework and guidelines provided by the IMF with China’s actual situation: 1.IMF points out that macroeconomic policies should be given priority when macroeconomic risks are the main consequences of capital flows.This paper finds that,(1)When cross-border capital flows mainly cause EMP risks,the policy should focus on foreign credit management,and in the face of depreciation pressure,we can choose increasing the interest rate spread and implement fullcaliber macro-prudential management;In the face of appreciation pressure,we can choose to expand the floating range of the exchange rate and relax the management of the central parity.The difference between China and IMF is that China should pay more attention to the coordination of macro-prudential policies,the expansion of exchange rate floating range and the management of exchange rate central parity;(2)When cross-border capital flows mainly cause money creation risks,the policy should focus on the management of the central parity,and coordinates with the management of foreign exchange credit,which is basically consistent with the IMF framework.However,China needs to pay more attention to the management of foreign exchange credit under extreme circumstances.2.The IMF argues that macroprudential policies should be used preferentially when the main consequence of capital flows is financial risk.This paper finds that when cross-border capital flows mainly cause risks to the banking sector,the policy focus is on macro-prudential policies,such as foreign exchange credit management.And in the face of the risk of foreign exchange balance deficit,the central parity management can be selected,and the floating range of exchange rate can be widened.China needs to pay attention to the distinction between inflows and outflows,and rely on the exchange rate floating range and the central parity management to play its role.3.The results of this paper show that China’s operation of cross-border capital flow management not only follows international conventions,but also bases on China’s reality.The management principle is not limited to micro subjects but to guide in macro level,so as to avoid the tendency of capital control.