noseeum wrote:Everything is true about your hypothetical high school athlete, except for one thing. That athlete is American. The Japanese and Latin American athlete does not have that option. In order to have the football option, you need to be playing it from at least 9th grade on, and they don't.

In Latin America, it's all about baseball. Sure, maybe soccer, but it doesn't compare to baseball in terms of popularity. When was the last Dominican star soccer player playing in Europe? That's more for South America.

And yes, of course the effect of more money is always positive, but I'm arguing it's not nearly as positive over the last 20 years as the scouting and development. You say $33,000 25 years ago is not that much. The per capita income for the Dominican Republic in 2005 was $7500. I can't find it, but you can bet it was way less than that in 1980. So even at only $33k and rising, the marginal utility, as you say, was decreasing rapidly while the investment in scouting and development was increasing rapidly. That was silly money to Latin American ballplayers.

Regardless of where a player lives, rising salaries impact their incentives and decisions. That's the point you consistently dodge. What's true for that American player---that rising salaries provide a stronger incentive to choose American baseball over other opportunities---is true for Japanese and DR players, too. The opportunities vary and the institutions vary, but the incentives and its impact on decisions are similar.

I'm glad to see that we now agree that the effect of rising salaries for Latino player is not zero. I've never disagreed that the academies are not part of the explanation of the influx of foreign-born players. So, we don't disagree on that point either.

It seems the only disagreement is the relative size of the effects, where there is no good data.

In 1980, DR per capita income was about $2,000. Remember that a player's decision is not "become a major league player". It's become "a minor league player who has about a 5% chance of becoming a major leaguer." Minor league players, of course, make much less. Even today, the average minor league salary is about $1,000 a month. I can't find a good source for in 1980, but if they've changed at the same rate as overall wages and prices, that salary was about $500 a month (just for the season, of course).

So, the typical DR player's economic decision might be roughly summarized as "make $2,000 in the DR" or "Play 6 months in the US for 3,000 and work the other 6 months in the DR for $1,000 and get the 5% chance that in 5 or 6 years you'll make $33,000 and have a small chance of making a $150,000 (the average 1980 salary)". Now, that's not a bad deal at all. At minimum you make an extra $2,000 PLUS you get a lottery ticket that's worth at least about $1,500 (.05 times $33,000) and could be worth more.

But compare that to now. Now, the decision is "make 7500 in the DR" or "play 6 months in the US for "6,000 and work the other 6 months in the DR for $3,750" and get a 5% chance that in 5 or so years you'll make 325,000 and have a chance of making $2,900,000 (the average salary in 2005). Now, at minimum, you make about $2,500 more. Your lottery ticket, however, is worth ten times as much AT A MINIMUM.

In economic terms what we would want to do is determine the expected value of that decision, taking into account the probabilities of getting to the majors and making it to the average salary level or even better. This gets us into the risk aspects of the choice I mentioned.

I did a little calculation in Excel with the following assumptions. Starting in 1977 and 1995, a player spends 5 years in the minors, then 6 in the majors. The first 4 years in the majors is at the minimum, then they make the average salary the last two years. The probability for any given player of getting to the majors is assumed to be 5 percent. If you get to the majors, there's a 40% chance you get the average salary, rather than the minimum. I add up all those expected values for each year and discount them to the first year of the decision.

In 1977 the expected value of those 11 years in baseball for a DR player was about $25,000 or roughly $2,000 a year. In 1995, the expected value was $90,000 or roughly $8,000 a year. Nearly 90 percent of the 4 times larger expected value came from the rise in the average salary in baseball. The deal become markedly better for Latino players compared to 20 years prior to 1995.

So, again, without good data, I see no reason to dismiss the rising salaries as an important determinant. The changing structure of the draft and the rules regarding Latino players drove up demand for them from teams and they set up baseball academies. At the same time, rising baseball salaries made the expected return to Latino players increase 4 fold. Both mattered and we don't know how much.