"Early work on the effeodifications to the empirical methodology or adjustments to the underlying economic model. The first studies on this topic relied on variation in the flow of immigrants into metropolitan areas. Geographic variation can understate the true effect of immigration if immigrants choose to settle in areas that experience positive labor demand shocks or if immigrant arrivals encourage similarly-skilled natives to move elsewhere. Borjas (2003) analyzed the effect of immigration on national skill groups instead and found a sizeable negative effect of immigration on native workers. Examples of papers that propose modifications to the standard model include Lewis (2011), which argues that cities with larger inflows of unskilled immigrants were slower to adopt labor-saving technologies, thereby preserving the demand for low-skilled workers; Ottaviano and Peri (NBER version in 2006, published in 2012), which suggests that immigrants and natives are somewhat imperfectly substitutable in production, even within skill categories; and Peri and Sparber (2009), which contends that immigration increases total factor productivity by facilitating specialization across tasks (see also Peri (2012) and Ottaviano, Peri and Wright (2013)). The contemporary literature on the effect of immigration on the US economy is too large for us to cover in full here; for recent surveys, see Hanson (2009) and Kerr and Kerr (2011)."