Overview of training theory

Training theory is an important theory of human resources development at present. Since 1960s+1980s, due to the accelerated change of social and economic environment, human resources, financial resources, material resources and technical resources have become one of the important resources in an organization. Especially in the service-oriented economic environment, it is obvious that the training of personnel is closely related to the effectiveness of organization and operation.

General training refers to the training provided by enterprises that can bring externalities to other enterprises. Are enterprises willing to invest in this kind of external training? How to determine the investment level and corresponding expenses of general training? These problems have attracted the attention of western labor economists.

Early research and Becker's general training theory [1]

Enterprise training has always been a concern of economists. Economists in the drought period realize that under the condition of market economy, appropriate skills can be obtained through training, but they also point out the problems that enterprises may face when implementing on-the-job training. For example, Pigou, an economist, believes that because employees can use training skills to serve other employers, enterprises do not have enough incentive to invest in employee skills training. He suggested that the government provide living allowance for on-the-job training of enterprises.

Becker (1964) made an enlightening theoretical analysis of general training, which provided a theoretical basis for future enterprise training research. He believes that in a competitive labor market, the wage rate paid by any enterprise is determined by the marginal productivity of other enterprises, and complete general training will be useful to many enterprises, and all the marginal products of these enterprises can increase the same margin, so the wage rate and marginal product increment are exactly the same, and enterprises that provide such training will not get any benefits. Employees who participate in the training are the only beneficiaries. A wise enterprise can only provide general training if it requires employees to pay all the training fees. In the absence of credit constraints/liquidity constraints, employees can rationally choose the training level to obtain the maximum benefits, and the general training investment behavior of enterprises can also reach the social optimal level. The way to solve the low-level training investment caused by credit constraints/liquidity constraints is to improve the lending market, rather than direct subsidies, which is different from previous studies.

General training theory under the assumption of information asymmetry [1]

An implicit assumption in Becker's theory is that employees can quickly become competent for the corresponding work and play the greatest role after training. In other words, potential employers have all the information about employee training. Katz and Zidman believe that this assumption is unrealistic. They (1990) revised the hypothesis that training information is symmetrical in a perfectly competitive market, and thought that the asymmetry of training information would affect the general training behavior of enterprises.

In their view, the value of general training investment is reflected in two parts: the net present value Z of specific post investment and the value OV of training options. Among them, Z refers to the present value of future productivity increment generated by general training. OV refers to improving employees' ability to cope with unexpected shocks and changes through training. For example, general training can be used as the basis for further training of employees in the future; It can make employees have a variety of work abilities and play a substitute role in personnel changes in other positions. More importantly, training can improve employees' ability to accept new technologies and adapt to organizational changes. This potential value may be very important for enterprises. Relatively speaking, ov information is more difficult to obtain than Z, and it is impossible for external enterprises to obtain all the information of employees.

The value of employees is an increasing function for enterprises to master training information. Only by mastering all the information of general training, including the content and degree of training, can enterprises arrange their employees reasonably and realize their maximum value. For training enterprises, all information about employee training can be obtained through observation and understanding during training, and the value of training employees at this time is Vt. For external enterprises, this information was not available at first. Only after the employment relationship begins can we gradually get to know the employees through the investigation of their work performance. Let I represent the number of inspection periods, and Vi represents the value of employees in the first period. With the increase of I, enterprises get more and more information, and Vi is also increasing gradually. The most effective way for enterprises to evaluate whether employees have received training and the degree of training is to arrange employees in positions that need training. In order to avoid the risks (such as decreased productivity or even negative value) caused by matching errors (placing untrained employees in positions that need training), enterprises prefer to place them in positions that do not need training. This way of inspection is inefficient, time-consuming and costly, and the value Vi of employees in external enterprises is lower than that of training enterprises. On the other hand, trained employees are unwilling to accept work arrangements provided by external enterprises in the face of long-term running-in and low efficiency. The above two reasons limit the flow of employees between enterprises and ensure the effectiveness of enterprise training investment to a certain extent.

General training theory of wage compression [1]

Becker's research shows that only when the marginal product increment is greater than the wage rate increment can the enterprise obtain part of the benefits brought by training and recover the investment cost. Acemoglu and Pischke( 1998b, 1999) have studied the wage structure. They believe that the incompleteness of the labor market and some institutional problems lead to the distortion of the wage structure, which is the reason why enterprises participate in general training.

The following is an analysis of the general training theory of Acemoglu and Pischke's wage compression.

Assume that the general training level of employees is τ, and the marginal products of all enterprises are f(τ), F' (τ) >; 0。 In the incomplete labor market, it is assumed that the cost of employee job-hopping is △≥0, and the value of employee choosing an external enterprise is V(τ) = f(τ)? δ. For training enterprises, as long as the wage level is set as the value level of employees choosing external enterprises, that is, W 1(τ) = V(τ) = f(τ)? Δ, can you keep employees because time gets f(τ)? W 1(τ)=δ gain. In this case, △ can be considered to be due to incomplete factors in the market, and enterprises occupy some products produced by employees for free. Acemoglu and Pischke call this income "rent". It can be seen from the above analysis that the rent has nothing to do with the training level τ, that is, in Figure L, f(τ) and W 1(τ) are two parallel curves. This shows that under any training level, enterprises can get a certain rental, so enterprises are unwilling to train employees' skills. When △=0, the labor market is completely competitive, which is consistent with Becker's training theory.

Figure 1 training cost-benefit analysis

Figure 2 Effect of Efficiency Wage/Minimum Wage on Enterprise Training [2]

In reality, the higher the skill level of employees, the more difficult it is to change jobs. Need to consider the factors of training. Let's assume that the letter △ is the training level τ is △ (τ) and △' (τ) > 0。 According to the above analysis, the training enterprise will set the salary as W2(τ) = V(τ) = f(τ)? δ (τ), thus obtaining: δ (τ) = f (τ)? W2(τ). Derive from both sides and get δ' (τ) = f' (τ)? W'2(τ)>0, which shows that training can improve productivity faster than the salary level, which is equivalent to compressing the salary level of employees. As shown in figure 1, as the skill level rises, the rent δ (τ) also rises, and the enterprise will benefit. As long as the training cost is not too high, the enterprise is worth investing in training.

Acemoglu and Pischke made a further study on the causes of wage structure compression, which can be summarized as two aspects:

1。 There are transaction costs in the labor market. This mainly refers to the cost of finding and matching in the labor market. The re-matching of labor force is difficult for both enterprises and employees. Especially for employees, this kind of loss is great. In reality, in order to avoid this loss, enterprises negotiate wages with employees. Let the bargaining power of employees be β; Through bargaining, employees get βf(τ) salary; Enterprises get (1? β)f(τ) product. With the increase of τ, the income obtained by enterprises is also increasing, which urges enterprises to invest in training. Acemoglu( 1997) made a detailed analysis of this.

2。 Information asymmetry between training enterprises and external enterprises. First, the training information is asymmetric. Katz and Ziderman( 1990) and Chang and Wang( 1996) believe that the asymmetry of training information between training enterprises and external enterprises leads to the difference in the value of employees who do not return to training enterprises, which in turn affects the salary structure. Secondly, the information asymmetry of employee competence. Acemoglu and Pischke( 1998a) believe that the general training effect is complementary to the ability of employees. Through training, employees with strong ability can produce higher value, while employees with low ability can produce relatively lower value. Employees who can sleep are most likely to be fired. Job transfer is considered to be caused by employees' inability to adapt to job requirements. Strong employees will avoid job-hopping and their job choices will be constrained. Employers use this constraint to realize rental income by paying wages below the productivity level, so they are willing to provide general training.

(b) There are special institutional arrangements between enterprises and markets.

1. Efficiency pay. Acemoglu and Pischke( 1998b) believe that there is information asymmetry between employers and employees, and there is a positive correlation between employees' value and their real efforts. Efficiency salary is the salary paid to employees by enterprises to motivate them to work hard, which is higher than the market level. Efficiency wage has the functions of encouragement and restraint. In Figure 2, it is assumed that the efficiency wage level meeting the constraint conditions is W *, and the corresponding skill level is τ f ... Delta > due to the rent; 0, the salary that the enterprise chooses to pay is Maxf(τ)? Δ, W *. For employees whose skill level is lower than τf, the enterprise will raise their skill level to τf through training, thus obtaining the maximum rent △.

2. Minimum wage system. The impact of minimum wage on training is similar to that of efficiency wage. As can be seen from Figure 2, if W* represents the minimum wage, there is also rent △ >; 0, the enterprise will raise the skill level of employees whose productivity is lower than f(τf) to τf through training, and will get all rents δ = f (τ f)? W *, thus becoming the de facto residual requester of productivity increment.

3. Trade unions. Acemoglu and Pischke( 1998b) believe that trade unions participate in wage negotiations on behalf of employees, which leads to extensive bargaining between enterprises and trade unions. When the trade union knows that the salary structure will affect the training decision of the enterprise, it will take the initiative to reduce the salary in the negotiation, support the enterprise to invest in training, and ensure the maximum interests of both parties.

Regarding the training level and cost, Acemoglu and Pischke( 1998b) mainly analyzed the situation of non-cooperation and perfect competition. Suppose the cost function of training is C(τ), which satisfies c' (τ) >; 0，C *(τ)& lt； 0 and has the inverse function c? 1(τ)。 In the case of non-cooperation, the training fee shall be borne by one of the enterprises or employees. The training level is also determined by one party alone, and the decision of employees to choose training conforms to the real maximum utility, that is, W' (τ w) = C' (τ w); Enterprises choose training to maximize profits, that is, to obtain δ' (τ f) = c' (τ f). Because the training of enterprises and employees is completely replaceable, when τ w >: τf, the training level required by enterprises is higher than that required by enterprises, all training expenses will be borne by employees; when τ w < τf, enterprises will bear all expenses (as shown in Figure 2). With the further compression of wages, the expected training level of enterprises is improved, and the expected level of employees is reduced, so enterprises are more willing to invest.