程志富,胡昌生
杠桿交易限制下可轉(zhuǎn)債的交換期權(quán)定價(jià)模型
程志富,胡昌生
(武漢大學(xué)經(jīng)濟(jì)與管理學(xué)院,湖北 武漢,430072)
不同于標(biāo)準(zhǔn)期權(quán),可轉(zhuǎn)債轉(zhuǎn)股條款的實(shí)質(zhì)是一份交換期權(quán)。考慮了市場中的杠桿交易限制,利用遠(yuǎn)期風(fēng)險(xiǎn)中性測度原理和超復(fù)制方法,重新構(gòu)建了一個(gè)可轉(zhuǎn)債交換期權(quán)模型。結(jié)合2015年2月至2015年10月的可轉(zhuǎn)債市場數(shù)據(jù),通過比較基于套利限制的交換期權(quán)定價(jià)方法與基于無套利原理的標(biāo)準(zhǔn)期權(quán)定價(jià)方法的計(jì)算結(jié)果,發(fā)現(xiàn)前者能夠?yàn)榭赊D(zhuǎn)債的真實(shí)價(jià)值確定一個(gè)參考區(qū)間,而后者僅僅是前者的一個(gè)特例,且后者大致相當(dāng)于價(jià)值參考區(qū)間的下界。通過調(diào)整杠桿交易限制的參數(shù),發(fā)現(xiàn):一方面,隨著市場杠桿交易限制的增加,轉(zhuǎn)股權(quán)(從而轉(zhuǎn)債)的價(jià)值會由于其復(fù)制成本的上升而增加;另一方面,當(dāng)轉(zhuǎn)股權(quán)處于深度虛值或深度實(shí)值狀態(tài)時(shí),轉(zhuǎn)股權(quán)(從而轉(zhuǎn)債)的理論價(jià)值區(qū)間會縮小。這就為杠桿交易限制與轉(zhuǎn)債價(jià)格無套利區(qū)間之間所存在的正向關(guān)系,以及轉(zhuǎn)股權(quán)實(shí)/虛值程度與轉(zhuǎn)債價(jià)格無套利區(qū)間之間所存在的反向關(guān)系提供了理論上的依據(jù)。
可轉(zhuǎn)債定價(jià);交換期權(quán);杠桿交易限制;超復(fù)制
從2014年7月以來,A股市場經(jīng)歷了一輪加速上漲的牛市行情。在此期間,散戶投資者前所未有地使用了杠桿,高峰時(shí)期,有高達(dá)4萬億元的資金通過各類融資渠道進(jìn)入市場,杠桿資金占市場交易額的比重接近20%,遠(yuǎn)超美歐成熟市場水平[1]。盡管如此,市場對于各種交易的杠桿比例仍然存在嚴(yán)格的限制。以A股市場杠桿交易資金占比超過55%的融資融券為例,根據(jù)滬、深證券交易所《融資融券交易試點(diǎn)實(shí)施細(xì)則》的相關(guān)規(guī)定,融資融券業(yè)務(wù)的保證金比例都不得低于50%,融資融券的交易杠桿限制在1~3倍之間,并有嚴(yán)格的券種和風(fēng)控限制。此外,其他形式的杠桿交易同樣面臨不同程度的杠桿比例限制。
由于經(jīng)典的期權(quán)定價(jià)理論建立于買空和賣空交易不受限制的理想假設(shè)之上,因此,真實(shí)股票市場上的杠桿交易限制,勢必對股票衍生產(chǎn)品價(jià)格帶來系統(tǒng)性的沖擊,形成對經(jīng)典模型價(jià)格的偏離。鑒于目前國內(nèi)市場上尚未推出標(biāo)準(zhǔn)化的個(gè)股期權(quán)產(chǎn)品,本文將以可轉(zhuǎn)債為例,深入探究股市杠桿交易受限條件下的期權(quán)定價(jià)問題。
長期以來,人們都將可轉(zhuǎn)換債券(下簡稱“轉(zhuǎn)債”)中的轉(zhuǎn)股條款視為普通期權(quán)。事實(shí)上,該條款所賦予投資人的權(quán)利是一份資產(chǎn)交換期權(quán)(下簡稱“交換期權(quán)”),即以債券資產(chǎn)兌換成一定數(shù)量股票資產(chǎn)的權(quán)利。關(guān)于交換期權(quán)定價(jià)方面的研究,最早可以追溯到Margrabe[2],他通過假定所有資產(chǎn)都服從幾何布朗運(yùn)動,得到了歐式交換期權(quán)的一個(gè)閉式解。Finnerty[3]將Margrabe[2]模型中的短期無風(fēng)險(xiǎn)利率和公司信用利差拓展到動態(tài)變化的情形,并利用停時(shí)理論分析了轉(zhuǎn)債的價(jià)值。國內(nèi)學(xué)者王靜和王敏[4]利用集對分析理論對轉(zhuǎn)債建模,并考察了她們的模型與Black-Scholes[5]期權(quán)模型以及交換期權(quán)模型[2]定價(jià)效果的差異。程志富等[6]在所建立的模型中重點(diǎn)考慮了轉(zhuǎn)股權(quán)的美式期權(quán)性質(zhì)以及公告期長短對轉(zhuǎn)債價(jià)值的影響,并利用非線性的最小二乘蒙特卡羅模擬方法對模型進(jìn)行了數(shù)值求解。但是,上述研究存在兩大明顯的不足:
首先,如上文所述,從股票資產(chǎn)角度來看,現(xiàn)實(shí)市場中的杠桿交易往往面臨不同程度的限制,使得經(jīng)典期權(quán)模型[2][5]所依附的市場完備性前提不再成立[7][8]。交易所及其他監(jiān)管部門在交易費(fèi)用、保證金、杠桿率以及持倉限額等方面的規(guī)定,顯然會影響到利用股票等資產(chǎn)復(fù)制交換期權(quán)的成本,從而導(dǎo)致交換期權(quán)的實(shí)際價(jià)值偏離經(jīng)典模型所對應(yīng)的期權(quán)價(jià)值。為便于表述,下文將經(jīng)典交換期權(quán)模型[2]理論價(jià)值統(tǒng)稱為Margrabe價(jià)格。
其次,從債券資產(chǎn)角度來看,盡管轉(zhuǎn)債中債券部分與股票部分均可視為由公司所發(fā)行的某種證券產(chǎn)品,但這并不意味著債券價(jià)值也遵循類似于幾何布朗運(yùn)動的過程。Merton[9]最早在無約束的條件下對公司債券進(jìn)行了結(jié)構(gòu)化建模,然而,Huang和Huang[10]等的實(shí)證研究卻發(fā)現(xiàn)結(jié)構(gòu)化模型的估值會顯著高于債券市價(jià)。債券價(jià)格之所以不服從Margrabe[2]所假設(shè)的幾何布朗運(yùn)動,主要原因在于:一方面,債券的利率和信用利差都具有均值回歸特點(diǎn)[11];另一方面,債券價(jià)格本身也不同于普通股股價(jià),前者通常都要受到相對固定到期期限的約束[3]。
為此,本文試圖修正Margrabe[2]對債券和股票資產(chǎn)的不合理假設(shè),重新構(gòu)建轉(zhuǎn)債的交換期權(quán)模型。
由于轉(zhuǎn)債的期限跨度較長,需要對市場利率進(jìn)行動態(tài)刻畫。根據(jù)無套利的Hull-White[12]模型對市場無風(fēng)險(xiǎn)利率的動態(tài)描述,無風(fēng)險(xiǎn)利率滿足如下增量過程
由于要給交換期權(quán)定價(jià),我們需要構(gòu)造一個(gè)債券遠(yuǎn)期,并以它作為計(jì)價(jià)單位來實(shí)現(xiàn)。
利用伊藤引理,得到純債券遠(yuǎn)期的收益率在風(fēng)險(xiǎn)中性世界里所遵循的增量過程
當(dāng)投資組合面臨杠桿交易限制,即
聯(lián)立式(1)~(13),可推得非完備市場下轉(zhuǎn)股權(quán)的無套利價(jià)格上界
按照Ingersoll[18],以及Tsiveriotis和Fernandes[19]等學(xué)者的觀點(diǎn),轉(zhuǎn)債的價(jià)值等于純債券價(jià)值與轉(zhuǎn)股權(quán)價(jià)值之和。其中純債券的價(jià)值由式(4)給出,轉(zhuǎn)股權(quán)的價(jià)值也由式(14)可得,于是,在杠桿交易限制下,轉(zhuǎn)債的無套利價(jià)格上界為
圖1 轉(zhuǎn)債的市場價(jià)格與理論價(jià)值
Figure 1 Market price and theoretical value of convertible bonds
圖中橫軸代表樣本觀察值個(gè)數(shù):第1~19個(gè)為歌爾轉(zhuǎn)債;第20~38個(gè)為格力轉(zhuǎn)債;第39~57個(gè)為電氣轉(zhuǎn)債;第58~76個(gè)為吉視轉(zhuǎn)債;第77~95個(gè)為洛鉬轉(zhuǎn)債;第96~110個(gè)為航信轉(zhuǎn)債;第111~125個(gè)為天集轉(zhuǎn)債。為了從統(tǒng)計(jì)意義上驗(yàn)證上述理論價(jià)值區(qū)間與市場價(jià)格之間的關(guān)系,我們對市場價(jià)格與區(qū)間上下界大小關(guān)系分別進(jìn)行了單側(cè)檢驗(yàn),檢驗(yàn)結(jié)果如表1所示。
由于兩個(gè)單側(cè)檢驗(yàn)的p值都小于0.05,因而,我們可以在0.05的顯著性水平上拒絕市場價(jià)格低于區(qū)間下界、高于區(qū)間上界的零假設(shè)。換言之,轉(zhuǎn)債市場價(jià)格基本處于我們所確定的理論價(jià)值區(qū)間內(nèi),模型估計(jì)效果良好。
表1 對市場價(jià)格的單側(cè)假設(shè)檢驗(yàn)
注:上述檢驗(yàn)在Matlab2016b完成,**對應(yīng)于的顯著性水平。
圖2 歌爾轉(zhuǎn)債的轉(zhuǎn)股權(quán)隨期權(quán)價(jià)值狀態(tài)的變化
Figure 2 Changes in the value of convertible bonds with the option value
在傳統(tǒng)的期權(quán)或其他衍生產(chǎn)品定價(jià)模型中,通常假設(shè)其標(biāo)的物市場具有完備性,從而得到其唯一的無套利價(jià)格。然而,從國內(nèi)可轉(zhuǎn)債市場的實(shí)際情況來看,由于市場存在交易限制等因素,尤其在股票市場杠桿交易受限的條件下,可轉(zhuǎn)債價(jià)格并不收斂于經(jīng)典的Margrabe模型價(jià)格,而是大致位于一個(gè)無套利的價(jià)格區(qū)間內(nèi)。同時(shí),可轉(zhuǎn)債中的轉(zhuǎn)股權(quán)也不同于普通期權(quán),而是一份以債權(quán)資產(chǎn)交換股權(quán)資產(chǎn)的交換期權(quán)。根據(jù)轉(zhuǎn)債中的轉(zhuǎn)股權(quán)交換期權(quán)屬性,并考慮到債券價(jià)值變化特點(diǎn)以及現(xiàn)實(shí)市場存在的杠桿交易限制,本文重新構(gòu)建了一個(gè)轉(zhuǎn)債的交換期權(quán)模型。本文為投資者提供了轉(zhuǎn)債合理價(jià)格的一個(gè)參考區(qū)間,即:在正常情況下,轉(zhuǎn)債市場價(jià)格既不應(yīng)該經(jīng)常低于不存在杠桿交易限制時(shí)的轉(zhuǎn)債價(jià)值,也不應(yīng)該經(jīng)常高于存在杠桿交易限制時(shí)的轉(zhuǎn)債價(jià)值。這對于處于信息嚴(yán)重匱乏或者信息過載的市場上的投資者而言,無疑具有一定的現(xiàn)實(shí)意義。進(jìn)一步的研究還發(fā)現(xiàn),一方面,隨著市場杠桿交易限制的增加,轉(zhuǎn)股權(quán)(從而轉(zhuǎn)債)的價(jià)值由于其復(fù)制成本的上升而增加;另一方面,當(dāng)轉(zhuǎn)股權(quán)處于深度虛值或深度實(shí)值狀態(tài)時(shí),轉(zhuǎn)股權(quán)(從而轉(zhuǎn)債)的理論價(jià)值區(qū)間會縮小。
目前,中國A股在經(jīng)歷了2014年7月以來的一輪大的漲跌之后,監(jiān)管當(dāng)局對于及時(shí)有效控制地控制市場風(fēng)險(xiǎn)的意識逐漸強(qiáng)化,杠桿交易限制對于期權(quán)等衍生產(chǎn)品的價(jià)格的沖擊也將更加不容忽視??梢灶A(yù)見,隨著杠桿交易透明度的不斷增強(qiáng)、信息披露制度的落實(shí)以及業(yè)務(wù)的杠桿比例控制嚴(yán)格化,將杠桿交易限制因素納入到期權(quán)等衍生產(chǎn)品的定價(jià)模型中也將越來越具有必要性和可操作性。
由于本文樣本數(shù)太少,無法對最終的分析結(jié)論進(jìn)行大樣本的檢驗(yàn),因此其可靠性還有待市場的進(jìn)一步驗(yàn)證。此外,模型忽略了轉(zhuǎn)股價(jià)修正條款,因此對于修正條款容易觸發(fā)且公司實(shí)施該條款的既定程序較為簡單的情形,本文的模型及有關(guān)結(jié)論可能不適用。當(dāng)然,這些也為本課題后續(xù)的進(jìn)一步研究提供了機(jī)遇和方向。
[1] 赫鳳杰. A股市場杠桿交易與監(jiān)管[J]. 財(cái)經(jīng)科學(xué),2015,(11):12-22.
[2] Margrabe W. The value of an option to exchange one asset for another[J]. Journal of Finance, 1978, 33(1):177-186.
[3] Finnerty J D. Valuing convertible bonds and the option to exchange bonds for stock[J]. Journal of Corporate Finance, 2015, (31): 91-115.
[4] 王靜,王敏.基于集對論的可轉(zhuǎn)換債券定價(jià)模型[J].商業(yè)研究,2008,(4):1-8.
Wang J, Wang M. SPA-Based Investigation on Pricing Model of Convertible Bonds[J].Commercial Research, 2008, (4):1-8.
[5] Black F, Scholes M. The Pricing of Options and Corporate Liabilities[J]. Journal of Political Economy, 1973, 81(3):637-654.
[6] 程志富,林勇,李巍.中國可轉(zhuǎn)債的美式交換期權(quán)定價(jià)與實(shí)證研究-基于非線性最小二乘蒙特卡洛模擬方法[J].武漢金融,2013,(1):30-33.
Chen Z F, Lin Y, Li W.Pricing and empirical research on American convertible options of convertible bonds in China -- Based on Nonlinear Least Squares Monte Carlo simulation (NLS-MC) [J].Wuhan Finance Monthly, 2013, (1):30-33.
[7] Finnerty, J D, Emery, Douglas R. Options in structured notes: fix the price or fix the spread? Fordham University Schools of Business Research Paper, 2014: No. 2493490.
[8] 胡昌生,程志富,陳晶,等. 投資組合約束下分離交易可轉(zhuǎn)債的定價(jià)[J]. 中國管理科學(xué),2017,25(09):11-18.
Hu C S, Cheng Z F, Chen J et al.Pricing Warrants Bonds under Portfolio Constraints[J]. Chinese Journal of Management Science, 2017, 25(09):11-18.
[9] Merton R C. On the Pricing of Corporate Debt: The Risk Structure of Interest Rates[J]. Journal of Finance, 1974, 29(2):449-470.
[10] Huang J, Huang M. How much of the Corporate-treasury yield spread is due to credit risk[J]. Working paper,2003.
[11] Karatzas I, Kou S. On the pricing of contingent claims under constraints[J].The Annuals of Applied Probability,1996,6:321-369.
[12] Hull J, White A. Pricing interest-rate derivative securities, Review of Financial Studies, 1990, 3(4): 573-592.
[13] Bhattacharya M. Convertible securities: their structures, valuation, and trading. In: Fabozzi, Frank J. (8th Ed.), The Handbook of Fixed Income Securities, Eighthed. McGraw-Hill, New York, NY, 2012: 899–945.
[14] Duffie D. Dynamic asset pricing theory. Princeton University Press, Princeton, NJ, 1992: chapter 7.
[15] Broadie M, Cvitanic J, Soner H M. On the cost of superreplication with transaction costs[J]. Review of Financial Studies, 1998, (11): 59-79.
[16] Cox AMG, Hobson D G. Local martingales, bubbles and option prices[J]. Finance and Stochastics, 2005, 9(4): 477-492.
[17] Schmock U, Shreve S E, Wystup U. Valuation of exotic options under shortselling constraints[J]. Finance and Stochastics, 2002, 6(2): 143-172.
[18] Ingersoll J E. A contingent-claims valuation of convertible securities[J]. Journal of Financial Economics, 1977, 4(3):289-321.
[19] Tsiveriotis K, Fernandes C. Valuing Convertible Bonds with Credit Risk[J]. Journal of Fixed Income, 1998, 8(2):95-102.
Exchange option pricing model of convertible bonds under leveraged transactions constraints
CHENG Zhifu, HU Changsheng
(Economics and Management School of Wuhan University, Wuhan 430072, China)
Convertible bond has been one of the innovative financial instruments in the investment and financing markets in recent decades. To study the law of value of convertible bonds, it is necessary to consider the objective characteristics of market development, but also grasp the essence of convertible bonds.
Leverage trading is a new thing for the A-share market. Since July 2014, its scale has soared, intensified the fluctuation of the market and led to the boom and slump of the recent market. It has made leveraged transactions constraints a priority for capital market regulation. At present, there are some incomplete cases in the actual market economy in China. Among those, leveraged transactions constraints are a typical reality in the current underlying market.
Besides, some literature combines theory and practice in real life to compare the converting option of the convertible bond and the common stock option and find the former is an exchange option which is equivalent of a right to exchange stock with a bond, while the latter is a vanilla option in substance. The convertible exchange option model is rebuilt based on the margin trading in the A-share market as well as the particularity of the conversion option,
The traditional approach to model convertible bond is Tsiveriotis and Fernande(1998) with a finite difference method or the pricing theory of Black and Scholes(1973) model. This approach separates the value of convertible bond into two parts: cash part and equity part, and it is made with no-arbitrage theory. Since many assumptions of classical models do not reflect actual situations in the real world, we modify the value model according to leveraged transactions constraints in the present to modify option pricing function on the arbitrage-free condition. In this paper, using the forward risk-neutral measure and super-replication method, the exchange option pricing model of convertible bonds is built under leveraged transactions constraints.
This paper empirically validates the exchange option pricing model by comparing Tsiveriotis and Fernande (1998) model and market prices for a sample of all corporate convertible bonds in the market from February 2015 to October 2015. The paper discusses and analyzes the impact of the leveraged transactions constraints on the theoretical and empirical values of convertible bonds.
Our empirical study shows that most market prices of convertible bonds in the Chinese market have been between the lower bound and upper bound of a no-arbitrage price interval from our model, and the value of convertible bond based on traditional model, a special case of the former, is almost equal to the lower bound of price. Meanwhile, the tighter leveraged transactions constraints in the market, or the less deep-in-the-money (deep-out-of-the-money) of a call option are, the larger arbitrage-free price interval of the convertible bond is.
As leveraged transactions constraints exist in reality, it is meaningful to build such a structure model. It shows that the model not only can reflect leveraged transactions constraints in the markets but also can capture the real-time changes on the depth of moneyness of converting option very well.
Convertible bonds pricing; Exchange option; Leveraged transactions constraints; Super-replication
2017-05-23
2017-11-18
Supported by the National Natural Science Foundation of China (71671134) and the Youth Project of the National Natural Science Foundation of China (71401128)
F830.91
A
1004-6062(2020)01-0195-005
10.13587/j.cnki.jieem.2020.01.021
2017-05-23
2017-11-18
國家自然科學(xué)基金資助項(xiàng)目(71671134);國家自然科學(xué)基金資助青年項(xiàng)目(71401128)
程志富(1986—),男,湖南株洲人;武漢大學(xué)經(jīng)濟(jì)與管理學(xué)院博士研究生;主要從事行為金融與資產(chǎn)定價(jià)研究。
中文編輯:杜 健;英文編輯:Charlie C. Chen