Matthew M. Singer, "Economic Voting in an Era of Non-Crisis: The Changing Electoral Agenda in Latin America, 1982-2010." Comparative Politics 45, 2 (January 2013): 169-185.
Abstract (article is gated).
Latin America's political economy has shifted in the three decades since the return to democratization, with the inflationary crises of the 1980s fading into the past. One consequence of this change is a reduction in the electoral salience of inflation. While electoral support for the incumbent in the 1980s and 1990s was strongly tied to his or her ability to prevent increases in prices, in the 2000-2010 period there was no significant association between inflation rates and election outcomes. Instead, incumbents who presided over a growing economy in the last decade reaped electoral benefits. The importance of the economy for electoral outcomes varies over time and even across economic indicators.
Basically, once inflation was tamed then voters began to care less about it. Moreover, they were willing to suffer relatively high rates of inflation (relative to history, that is) as long as they felt the government was promoting growth sufficiently (e.g. countries like Argentina and Venezuela).
Another angle to take is political learning to help explain why inflation is much less of a problem than in the past. Presidents have seen how badly their predecessors were punished for inflation, and therefore take greater strides to prevent it. It's not clear where the "inflation punishment threshold" is. In other words, at point is growth not enough because voters see their purchasing power diminishing too much?