It's a bit surreal when an academic study reads like a biography.

Whenever a new study on the decline of driving in America is released, it's almost like reading a chapter of my life and the people around me.

Three years ago, I lived in rural New England and drove my vehicle to get to most places -- particularly in the winter. But like a lot of Millennials, I moved to Washington, D.C. after the recession into a booming local economy.

I quickly ditched my car for a number of reasons: easy access to public transportation, high auto insurance rates in D.C., and the general hassle of owning a car in a city. There's also the thrill of riding a bicycle through gridlocked traffic, a feeling that any urban bicyclist will describe with glee.

I'm certainly not alone. According to research from the Public Interest Research Group, young Americans between the ages of 16 and 34 are driving 23 percent less than they did in 2001. All Americans are driving less, but the decline is even steeper for Millennials. 

In study after study, the trend is stark. But researchers are still trying to figure out whether the decline in driving is due to a post-recession hangover, or caused by structural long-term changes that mean "peak car" has arrived.

In his latest of a series of studies on the issue, Michael Sivak of the University of Michigan Transportation Research Institute doesn't definitively conclude that America has reached its driving peak. But he adds yet more compelling data to the mix hinting at a long-term trend.

In two earlier studies from earlier this year, Sivak reported that light-duty vehicles per person, per licensed driver and per household had all peaked before the economic turmoil in 2008, suggesting that "other societal changes" like public transportation, urbanization and telecommuting were core drivers of the trend.

Sivak's most recent research adds fuel consumption to the mix. He found that fuel consumption rates per person, per household and per vehicle all peaked in 2004, well before the economy tanked. Sivak also noted that fuel consumption is dropping faster than miles driven, which shows that fuel efficiency standards are working.

"The findings of the present study indicate that the corresponding rates for fuel consumed also reached their maxima during [2003-2004]. Thus, the combined evidence from these three studies indicates that -- per person, per driver, and per household -- we now have fewer light-duty vehicles, we drive each of them less, and we consume less fuel than in the past," concluded Sivak.

Here's what that downward trend in vehicle mileage looks like:

This means that fuel consumption in 2011 was lower than in the mid-1980s.

In 1984, per-person fuel consumption was 400 gallons per year. In 2011, it was 398 gallons.

In 1984, fuel consumption per licensed driver was 608 gallons. In 2011, it was 585 gallons.

In 1984, households consumed 1,106 gallons of gas each year. In 2011, they consumed 1,033 gallons.

These factors certainly make a "peak car" scenario more compelling. But it all depends on how that peak is defined. Sivak concludes that fuel consumption rates will continue to fall; however, population growth and an economic rebound will likely mean an increase in total miles driven. If the decline in fuel consumption rates outpace the increase in vehicle miles driven, then one could argue that a peak has occurred. A true peak would be a concurrent drop in vehicle miles traveled and fuel consumption.

Sivak and others studying the issue aren't ready to make a definitive conclusion yet. We may have to wait a few more years to separate the structural changes from the lasting effects of the economic downturn.

"Because the changes in the rates from 2008 on likely reflect both the relevant societal changes and the current economic downturn, whether the recent maxima in the rates will represent long-term peaks as well will be determined by the extent to which the relevant societal changes turn out to be permanent," wrote Sivak.