Judge Richard Posner (who's considered "conservative") has this to say on the subject:
Increasing income taxes can . . . reduce the amount of work. But when the issue is increasing marginal income tax rates, the likelihood of a significant effect on work depends critically not only on the amount of the increase but also on where the margin is set. The Obama Administration proposes by allowing some of the Bush tax cuts to expire on schedule to increase the marginal income tax rate of persons whose taxable income exceeds $250,000 a year from 35 to 39.6 percent. . . . The effect on most of these taxpayers would be small. Suppose a person earning $300,000 in taxable income pays $75,000 in taxes currently. If his marginal rate (the rate for taxable income above $250,000) rises from 35 to 39.6 percent (for simplicity I’ll round this up to 40 percent), then his total income tax bill will rise from $75,000 to $77,500 (because on the last $50,000 of his income he will pay an extra 5 percent in tax). Will this increase in his total federal income tax bill, of 3.3 percent, cause him to work less? I would like to see evidence that it would.A Berkeley economics professor, Owen Zidar, has studied the macroeconomic effects of tax changes on different income groups by looking at decades of data on U.S. federal and state taxes (respectively, from 1945-2010 and 1980-2011). (Here's the abstract with a link to download the whole paper.) He concludes:
Because of tax avoidance opportunities, it seems that very few people pay more than about a quarter of their income in federal income tax. Suppose someone has an income of $10 million on which he currently pays $2.5 million in income tax. The Administration’s tax increase would raise his income tax by $487,500 ([$10 million - $250,000] x .05), or slightly less than 20 percent. Would that affect how hard he works? I’m skeptical.
Income tax on earned income . . . does increase the cost of work relative to leisure, which is the basis for concern that increasing income tax rates, especially marginal rates, will cause a shift from work to leisure (and . . . to nonpecuniary work such as household production). But there is another effect: by lowering disposable income, an increase in income tax may cause a person to work harder in order to maintain his previous standard of living. The net effect on income of a higher tax is therefore uncertain. . . .
Americans are accustomed to working hard and to achieving and maintaining a high level of expenditure on consumer goods. In addition, people who earn high incomes tend to be highly competitive and to rate themselves by their income relative to the income of peers. Because of the absence of an aristocracy and the presence of a generally philistine attitude toward culture, money is the key index to prestige and social standing in the United States and, for many Americans, to a feeling of self-worth. These features of American society lead to me to be skeptical in general about the effect on work and therefore income of increasing marginal income tax rates to the levels contempated by the Administration.
Do tax cuts that go to high income taxpayers generate more output and employment growth than similarly sized tax cuts for low and moderate income taxpayers? . . .(You might object that this kind of study is skewed by the fact that taxes are sometimes raised or cut in response to specific economic conditions. But Professor Zidar tried to exclude those kinds of tax changes from his study.)
If tax cuts for high-income earners generate substantial economic activity and job creation, then we should expect to see three things in the data. First, employment growth should tend to be higher in the years following exogenous tax cuts for the rich. Second, places with a higher share of rich people should grow faster following national tax cuts for the rich (since these areas receive more tax cuts for rich people in dollar per capita terms). Similarly, growth should be lower following tax increases on the rich, especially in places where many rich people live. None of these predictions are born out out in the data.
I find that the relationship between upper income tax changes and growth is negligible in magnitude and substantially weaker than equivalently sized tax changes for the bottom 90%. The point estimates suggest that almost all of the stimulative effect of exogenous tax cuts is due to tax cuts for the bottom 90%. Differential consumption responses help explain why a dollar of tax cuts for the top 10% produces less growth than those for the bottom 90%. Investment responses are also stronger following tax cuts for the bottom 90%, suggesting that the effects of additional economic growth tend to exceed the effects from income changes among those who are more likely to save. Overall, tax cuts for the bottom 90% tend to result in more output, employment, consumption and investment growth than equivalently sized tax cuts for the top 10%.