The Core Features of Internet Finance

First, the Internet can replace traditional financial intermediaries and physical outlets and manual services in the market, thus reducing transaction costs. For example, mobile banking itself does not need to set up outlets and additional equipment and personnel, and the transaction cost is obviously lower than that of physical outlets and manual tellers (CGAP, 20 10).

Second, the Internet promotes operational optimization and can reduce transaction costs. For example, integrating multiple bank accounts with third-party payment can improve the efficiency of payment settlement. Under the traditional payment mode, customers can't establish contact with the central bank directly, but must establish contact with commercial banks separately. In the third-party payment model, customers establish contact with third-party payment companies, and third-party payment companies establish contact with commercial banks, not customers. At this time, the third-party payment company becomes the counterparty between the customer and the commercial bank, and a large number of small transactions are settled by secondary settlement, which reduces the transaction cost (Xie, 20 14a).

Thirdly, the disintermediation trend of internet finance shortens the financing chain (which will be discussed in detail later), which can reduce transaction costs.

Second, the degree of information asymmetry is reduced.

In internet finance, big data is widely used in information processing (embodied in various algorithms, automation, high-speed and network operation), which improves the efficiency of risk pricing and risk management and significantly reduces information asymmetry.

There is no unified definition of big data so far. However, it is generally believed that big data has four basic characteristics-huge data volume, low value density (some people understand that it is of great application value), wide sources, many types and fast growth (fast speed, some people understand that it needs high-speed analysis ability). The background of big data is the digitalization of the whole society, especially the development of social networks and various sensing devices (see the first part). There are three main types of big data-recorded data, chart-based data and ordered data. With the development of cloud computing and search engines, it is possible to analyze big data efficiently. The core problem is how to quickly extract price information from a wide variety of huge data, and the main task is to solve the problem (area, group, 00; Oil a, a02.t.03) chapter-moving. The task is to predict some specific health based on some sexual values. The second category is the task of emotional narration, whose goal is to derive a model to summarize the potential connections in historical evidence. In the process of entry, general affairs, fruits, my traces and anomalies can be divided into scores, regression and Zheng points. Cluster analysis, recommendation system, anomaly detection, energy correlation analysis, etc. Big data analysis is pragmatic. Prediction occupies a large proportion in big data analysis, and post-evaluation of prediction results is also an important content of big data analysis. The combination of big data and ultra-high-speed computers will make correlation analysis more important than causal analysis, and behavior analysis will be no less important than financial statement analysis.

In the field of credit, the dynamic default probability can be determined according to big data. Xie Ping and Zou Chuanwei (20 12) pointed out that many stakeholders can evaluate a credit subject on the Internet, so that the default probability can be known at any time based on independent information and subjective judgment, which is the most effective. The overall effect is that both local information and private information are public, and information can only be explicit, decentralized and centralized. Indicators or indexes such as "sufficient statistics" can reflect the collected information and make it "equal" between people. We call this situation "public comment" principle for short, which can replace professional and linear credit evaluation methods in banks.

The securities market may have the characteristics described by behavioral finance (Shefrin and Statman, 1994) and efficient market hypothesis (Fama et al., 1969). On the one hand, driven by social networks, communication, interaction and mutual influence among investors will be very effective, and individual and group behaviors will approach the description of behavioral finance (for example, Coviello etal .(20 14) found that human emotions can produce contagious effects through social networks, and then have an observable impact on individual securities or the entire securities market. On the other hand, with the popularization of big data analysis (insider information does not belong to big data), market information is sufficient and transparent, and market pricing efficiency is high (for example, some complex calculations in securities pricing are transformed into applications and simplified), the securities market will approach the description of efficient market hypothesis.

In the insurance field, big data can improve the accuracy of insurance actuarial, so that premiums can fully consider individual differences and be dynamically adjusted, similar to the dynamic default probability. For example, in non-life insurance, insurance companies can provide customers with insurance based on mileage and time (paying by driving behavior) and insurance based on driving behavior (paying by driving style) to help the insured improve their driving habits (managing driving style); On the basis of the life table, the life insurance actuary fully considers the individual's genes, family inheritance, diet and exercise habits, and the future occupation, and the timeliness is further improved (Wang He, 20 14). With the improvement of actuarial efficiency, insurance in Internet finance will approach a perfect risk transfer model-voluntary, free and fair risk transfer (ATow, 1970). First, with the enrichment of insurance products, there may be corresponding insurance products for each risk in terms of personal and property. Second, the premium rate is determined by the principle of fairness. Third, the risk is transferred to people with corresponding risk preferences in society, and they bear it themselves.

Third, the expansion of the transaction possibility set

The Internet has gradually reduced the transaction cost and information asymmetry, expanded the possibility of financial transactions, and made the transactions that were originally impossible possible (Xie Ping and Zou Chuanwei, 20 12). For example, in P2P network loans, strangers can also borrow money, while direct lending between offline individuals generally only happens between relatives and friends. In crowdfunding financing, the smoke of investors and fundraisers must have quality control, and the sooner they choose, the better. Please ask Jin Chao EV to add some from time to time. Use that line in the inspection. 2.m= Kaodu, expected period, more legal, state affairs. It should be pointed out that the combination of five products is also helpful to expand the transaction possibility set of Internet finance.

But the possibility of trading collectively treats the risk of "Zhang Zhang". First, the financial knowledge, risk identification and tolerance of online lottery players are relatively lacking. Second, the investment of these people is small and scattered, and the problem of "hitchhiking" is prominent, and the grid discipline for Internet prosperity is easy to fail. Third, individual irrationality and collective irrationality are more likely to appear and gather. Once there is a risk in internet finance, the amount of funds involved will have a highly negative impact on society. Therefore, the protection of financial consumers is an important part of internet financial supervision (Xie Ping, 20 14a).

Fourth, the transaction is disintermediated.

In internet finance, the matching of the term, quantity and risk of capital supply and demand does not necessarily need to go through traditional financial intermediaries and markets such as banks, securities companies and exchanges, but can be directly matched through the Internet.

In the field of credit, individuals and small and micro enterprises have endogenous loan demand in consumption, investment and production (such as smoothing consumption, starting investment projects and liquidity demand). These loan demands belong to legal rights (that is, loan rights. At the same time, it is also a legal right for individuals to bear their own risks to maintain and increase wealth through investment. However, these loan rights and investment rights are scattered, facing matching problems and transaction cost constraints. For example, the "two-plus dilemma" in many places in China (it is difficult for enterprises to raise more funds and invest more funds) reflects this friction in the credit field. P2P online lending can alleviate the information asymmetry in the matching of loan rights and investment rights, and reduce transaction costs, which is inevitable. Many loan rights and investment rights that traditional finance can't satisfy have been satisfied through P2P network loans. In places with good credit foundation (such as the United States), the vitality of P2P online lending has emerged. In addition, repeated blogs between P2P online lending platform and borrowers can curb fraud. In the context of big data, there is a positive incentive mechanism between financial democratization, popularization and data accumulation.

In the field of securities, under the current technical conditions, investors can directly open accounts in the stock exchange, without going through securities companies, realizing 100% online transactions, and the brokerage business of securities companies becomes unnecessary. In addition, a "financing toolbox" may appear. When the information is transparent enough and the transaction cost is low enough, some enterprises (especially those with good qualifications) can directly raise funds on crowdfunding platforms (even their own websites) without going through the stock market or bond market, and various financing methods are integrated. Enterprises dynamically issue stocks, bonds or mixed capital instruments according to their own needs for investors to choose. Investors can know their portfolio position, market value, dividends, maturity date and other information in real time, and can also transfer and trade securities with each other.

In the field of insurance, there will be a "full insurance" model (Wang He, 20 14). The core function of insurance is economic compensation, that is, insurance companies provide economic compensation for unexpected losses based on the theorem of large numbers. In economic compensation, the insured who has not suffered unexpected losses indirectly compensates the insured who has suffered unexpected losses through the premiums paid by himself. In the ideal situation of full competition, the premiums paid by all the insured should just cover their overall unexpected losses (that is, the principle of net balance). The person who plays the role of premium transfer among insurance companies can sign the agreement through the grid, identify the work station and open the risk body as soon as possible. In fact, the bureau is obliged to give compensation to help each other, which is worse than bribing people. One country, one product, collects Zhang Wanming's commune through the original platform, such as the "Anti-Daoism Commune" in the city country. One version "Members provide 0 yuan's help, collect members from the surface, set up a business, and try to release a person." The big bang technology makes information more and more like a function. The role of payment. The "all-insurance" model embodies the disintermediation of insurance.