In general, owners of capital want to make it available for usage, because productivity-enhancing capital tends to be more valuable if there's labor involved whose productivity can be enhanced. To oversimplify, productivity = capital * labor, so capital owners have every incentive to invest in both capital and labor.

There's quite a bit of empirical research that indicates that across a wide range of political regimes, wages track very closely with labor productivity. When technological advances or increases in industrial development make labor more productive, demand for labor increases, which bids up wages.

You are correct that full employment is not guaranteed. However, if you look at top-tier industrialized countries today, countries with relatively libertarian economic policies tend to have lower unemployment rates. Today's 9-10% unemployment in the US is the depths of the worst recession in decades, but it's not far from normal non-recession employment levels in much of Western Europe.

Government unemployment benefits at best have no significant impact on the total productive capacity of society, as someone who is unemployed is (by definition) not working whether they're collecting unemployment checks or not. What unemployment benefits do do is improve the standard of living of the unemployed by taking a slice of the income of everyone who is currently employed and reallocating it.

In a radically libertarian society, I'd expect to see very low unemployment and relativily high real wages for the employed, but the unemployed would be dependant on their own savings and private charity (whether organized charity or gifts from family and friends) to tide themselves over during bad times.

(Reply) (Parent) (Thread)

Yes, wages do tend to track productivity (although to nitpick, labor costs track productivity; wages do not, due to the increasing share of non-wage compensation).

Unemployment benefits indeed do not have a significant impact on productivity, nor (for that matter) impose a significant tax on the employed, simply by virtue of the relatively small amount spent. The unemployment check is, however, often of immense importance to the unemployed individuals involved. Even small amounts of benefits matter at that level.

I do not think that more libertarian policies lead to high employment, if only because libertarianism here is a broad brush that entails all of low taxes, low state provision of services, high state protection of property rights, and low state macroeconomic intervention - while some of elements of each these promote higher employment, some of these definitely militate against it. In correlating indices of economic freedom against desirable outcomes (life expectancy, growth, political freedom, etc.), good government indices matter more than small government indices.

Although of course, if you think radically libertarian policies do generate high employment, then you can conclude that radically libertarian policies generate high employment. ;p

(Reply) (Parent) (Thread)

Of course, since we're talking about a radically libertarian society, we have no data for what a modern developed economy would look like because it's not something that's been tried. Post-WW2, developed countries range in economic libertarianness from Signapore to Sweden, which is not an enourmous range compared to the entire sweep of human societies. So we're either dealing entirely with theory (in which case there are huge legitimate disagreements and a high level of uncertaintly), or we're extrapolated far beyond the data we have (either we limit ourselves to modern developed countries and extrapolate far past the Singapore end, or we include third-world countries in our dataset and try to figure out how to score countries that have low levels of taxation and government services but which also have weak state protection of property rights, high levels of regime risk, and arbitrary enforcement of regulations and contract law).

(Reply) (Parent) (Thread)

Also, I fully accept your caveat that it's cost-of-compensation that tracks productivity, not cash wages. I should have spoken more precisely.

(Reply) (Parent) (Thread)

Wages haven't tracked productivity gains in 10 years in the US. There is quite a bit of actual real world data on this.

Between 1947 and 1974, productivity, or output per hour, and median family income, adjusted for inflation, both roughly doubled. Between 1974 and 2000, productivity rose 56% while income rose 29%. Between 2000 and 2005, productivity rose 16% while median income fell 2%.

So much for that theory.

(Reply) (Parent) (Thread)

That's an artifact of how the two indexes are calculated. There's two issues. First, the disparity between cash wages and cost of total compensation that dnwq brought up, as most employers have been spending more on non-cash benefits over that time period. Second, the labor productivity index and wage statistics are adjusted for inflation using different measurements of inflation. Labor productivity is adjusted using the NFB deflator (measures inflation in raw material and equipment costs for non-farm businesses), while wages are adjusted using CPI-U (measures inflation in prices of goods purchased by urban consumers). Between 2000 and 2005, energy prices went way up, and energy prices are weighted much more heavily in CPI-U than in the NFB deflator.

If you use total compensation package numbers instead of cash wages, and you either use nominal numbers for both wages and productivity or you correct for inflation using either CPI-U for both or NFB-deflator for both, then the apparent productivity gap goes away almost completely.

Source:
http://www.econbrowser.com/archives/2006/09/productivity_an.html

Scroll down to "Late addition: 19 September, 6:10pm Pacific".

(Reply) (Parent) (Thread)