From the first chapter to the third chapter, this book describes and explains the success and failure of American economic development. First of all, increasing inequality relative to other advanced countries requires an institutional explanation other than economic supply and demand theory. The low degree of American unionization and formal training matters. However, the solution is not so simple. There are trade-offs between different institutional wage setting and training mechanisms.

* Much of the evidence and arguments across three chapters is redundant. Don’t spend much time to read in detail.

Ch. 1. How Labor Fares in Advanced Economics - Richard B. Freeman
This chapter shows evidence of the falling position of American workers vis-a-vis those in other advanced countries. In the trade-off between creation of jobs and wage level, the U.S. chose the former and traded low wages for unskilled workers for low unemployment: much U.S. job growth occurred in low-wage service industries while many women have been pushed into the job market by the falling income of their husbands (p5t). Because America shows higher mobility and greater flexibility in hiring, firing, or altering work conditions than Europe, people move rapidly between unemployment and employment. However, this high mobility or flexibility implies a high cost of finding and training new employees.
In terms of productivity and living standards, America is doing well. But her productivity is growing less rapidly than that of other countries, which implies that the U.S. position is likely to worsen by the year 2000 (p10m). The most conspicuously inferior aspect of American performance appears in the low wages or great poverty of workers. Although the U.K. also shows increasing inequality since the 1980s, the U.S. is the only country that shows  decreasing real wages of low paid workers (p39m). In sum, while the United States did well in job expansion, it did poorly in earnings growth.

Differences in rules and institutions affect economic outcomes and are themselves an important element in working life. The U.S. has a business-oriented union movement based largely on relatively autonomous local unions and employers. The process of bargaining between employees and employers is not collective on the national level, but individual within local situations. This is a big contrast with European countries where unions' and employers’ federations negotiate collectively as social partners (Fig. 1.5). As a consequence, the U.S. represents a decentralized extreme in wage setting among employers, employees, and governments, which leads to easy adaptation to specific environments but high wage differentiation.  On the other hand, all advanced countries experienced some decentralization in the 1980s.
In addition, the U.S. mandates fewer rights and benefits for workers than do European countries. Note that, however, the high benefits and job security of European workers are costly. In terms of state welfare, the greater the social wage regardless of their work status, the smaller will be the need to work and the higher the wage that people will demand to work. Anyway, the United States may have something to learn from others.
 
Ch. 2.  Rising Wage Inequality: The United States vs. Other Advanced Countries
- R.B. Freeman and Lawrence F. Katz
(1) The most decentralized labor markets and wage-setting systems—U.S. and U.K.—had exceptional increases in earnings inequality during the 1980.
(2) One of the causes was a shift in relative demand that favored more educated workers with problem solving skills. This increasing demand was driven by technological innovation.
(3) The demand side explanation is incomplete because the increasing demand for skilled labor is similar across advanced countries. Slower growth in the supply of highly educated workers in the U.S. contributed to the educational wage differentials. Such slow growth of supply was largely due to a slower expansion of the college-educated work force and the influx of low educated immigrants in the 1980s.
(4) The supply and demand explanation is still incomplete. Institutions matter: the decline of unionization, and weak training for low-educated workers. U.S. workers did not adapt to technological change well.

The authors name factors (2) to (4) the SDI (supply, demand, and institution) explanation of increasing inequality in the U.S. Though the authors point out institutional factors, they also mention their explanatory limit. Institutions are not immune to market forces. Unionization fell in most countries in the 1980s. Clearly, there is a space for institutions to affect outcomes, but only within limits. Keeping the limitation in mind, it seems desirable to buffer the earnings of the less educated with institutional wage setting while increasing training to build their skills.

Ch. 3.  Payoffs to Alternative Training Strategies at Work - Lisa M. Lynch
There is a dilemma from the employer side in training employees when a firm is small or the labor market shows high employee turnover.
(1) Japan solves it by imposing high costs on employees who leave a firm (p65m).
(2) German apprenticeship training is characterized by coinvestment and codetermination by workers and firms aided by national certification of skills. German firms undertake general skills training (vs. U.S. task-specific training) which generates a high-skill, high-productivity labor force with high adaptability to new technological environments.
(3) Sweden and other Nordic countries have government-sponsored programs, whose effect is questionable.
(4) The U.S. training system is highly decentralized and has little formal structure. No national system exists. Decisions to invest in training are made by workers themselves or by specific firms. In the workplace, workers learn just by doing. Most workers who receive employer-provided training are technical and managerial employees with a university degree (p71b).

Through comparative and quantitative data, it is confirmed that a workplace-based skill training system or apprenticeship is more effective than a school-based or government-led system, and much more effective than informal training (pp85-86). In addition, noting that U.S. training expenditures are comparable to those of other developed countries, it is significant that U.S. and U.K. firm-specific training failed in preparing for technical change (p76m-b). By contrast, youths in Germany are willing to work hard in school to get better apprenticeships, and to accept lower wages during the apprenticeship. The lower wages allow firms to provide more general training, creating a virtuous circle. In other words, general training can be provided by private firms when there is a accepted national system to recognize and certify general skills which helps workers to accept low wages. It is also noteworthy that unions push for more general training.

Author’s best answer for training policy: governmental institutional support to overcome the potential market failure inherent in firm-based general training (p88t). For example: apprenticeship training that includes nationally recognized certification of skills and an educational system of workplace relevance.