The answer is "No" (although earthquakes are different). A new paper out in Natural Hazards Review adds additional empirical support for this conclusion (Tansel and Sizirici 2011). Rather than improving, there is evidence that in some places more recent construction actually performs worse in hurricanes than does older structures. There is also evidence that recent hurricane experience is associated with better construction.
The paper looked at damage to buildings following Hurricane Andrew (1992) and compared the scope of damage to the year of construction (results shown in the figure above). Here is an except of the paper's conclusions:
Southeast Florida has experienced a significant increase in population density and urbanization since the 1950s. In this study, field data were used to analyze the damage to the housing units based on wind speed, age of buildings, distance to coast, and distance to the hurricane eye. A relatively long period of low hurricane activity (1950 to 1992) has resulted in increased population density and coastal development in South Florida, significantly increasing the vulnerability of the region. The structural damage data compiled by the Metropolitan Dade Department of Building and Zoning after Hurricane Andrew in 1992 show that the long period of low hurricane activity in Dade County during 1950–1990 resulted in lower construction quality. In general, the older housing units constructed after the previous major hurricanes (in 1926, 1945, and 1950) showed less damage than the units constructed during 1970–1990.Of course, our normalized hurricane losses show no trend which matches up perfectly with the lack of trend in hurricane frequency or intensity at landfall, so there is no evidence of a bias one way or the other. However, building performance in hurricanes provides a good example that things do not always get better all the time.