The use of nonpharmaceutical interventions (NPIs) to slow disease spread, is a part of national pandemic preparedness as the first line of defense against influenza pandemics. Preemptive school closures (PSCs), an NPI reserved for use in severe pandemics, are highly effective in slowing influenza spread but have unintended consequences. We simulated PSC impacts during a 1957–like pandemic to estimate population impacts and quantify costs of closing schools at the national level. We also simulated 1957–like, 1968–like, and 2009–like pandemics at the Chicago regional level. We estimated economic impacts resulting from loss of income due to illness, providing childcare during closures, and other PSC costs while taking into consideration the number of cases averted with each mitigation strategy. The estimated net PSC costs ranged from $15 billion to $192 billion (2016 USD) (1957–like, national level) where between 2.3 and 47 million US cases may be averted depending on strategy. We found that 2–week school–by–school PSCs (as opposed to county–wide or school district–wide ones) had the lowest cost per discounted life–year gained for both national and Chicago regional–level analyses of all pandemics. While feasibility of such spatiotemporally precise triggering is presently questionable for most locales, this is, theoretically, an attractive option early in an outbreak, while assessing transmissibility and severity of a novel influenza virus. In contrast, we found that county–wide PSCs of longer durations (8 to 12 weeks) would result in the most averted cases (31–47 million) and deaths (105,000–156,000), albeit at considerably more cost ($125–$150 billion net of averted illness costs) for the national–level, 1957–like analysis. The estimated net costs per death averted ($1.0 to $1.2 million) for these scenarios compare favorably to the range of values recommended for regulatory impact analyses ($4.6 to 15.0 million). Hence, economic benefits of such PSCs would exceed the population impacts and economic costs.