In a December 2010 issue of The Wall Street Journal, Stanford’s Michael Boskin provided an extensive list of recent studies from Columbia, University of Chicago, Stanford, UC Berkeley, University of London, and others that support the notion that reduced taxes and decreased government spending bring about economic growth. Some have interpreted this as a movement toward a new economic consensus. The research of Harvard's Alberto Alesina and Silvia Ardagna on fiscal policies spanned 21 countries and roughly 40 years. Their findings indicate that economic expansion occurred under reduced taxes, while recessionary periods were under tax hikes. A 2010 study by Swedish economist Andreas Bergh and Oxford's Martin Karlsson found that increased government spending (surprise, surprise) reduces economic growth, even in Bergh's own country (the poster child of social democracy).
Economist Thomas Sowell writes,
When the tax rate on the highest incomes was 73 percent in 1921, that brought in less tax revenue than after the tax rate was cut to 24 percent in 1925. Why? Because high tax rates that people don't actually pay do not bring in as much hard cash as lower tax rates that they do pay...Then and now, people with the highest incomes have had the greatest flexibility as to where they will put their money. Buying tax-exempt bonds is just one of the many ways that "millionaires and billionaires" avoid paying hard cash to the government, no matter how high the tax rates go...Even more so today than in the 1920s, billions of dollars can be sent overseas electronically, almost instantaneously, to be invested in other countries--creating jobs there, while millions of Americans are unemployed...Despite political demagoguery about "tax cuts for the rich," in human terms the rich have less at stake than working people. Precisely because the rich have so many ways of avoiding taxes, a high tax rate is likely to do them far less harm than it does to the economy, on which millions of people depend for jobs.
Zakaria's foreign evidence (e.g.