An incentive that works less well than I thought

A recent study indicating that tax incentives for retirement savings are not very effective is changing my thinking about that incentive, specifically about how it's worked in my life.

My wife and I are savers. Arguably, we've had to be, as our income was pretty low until our late thirties, and neither of us is building a fixed pension.  Plus, we've had maybe 40-60% of our money in stock funds since we started saving in the late nineties, and what with two market crashes I reckon our return has been less than 5% per year.

Since the late '90s I've had the kind of solo retirement accounts allowed to the self-employed. Early on, I was allowed to save 20% of my income, and I was comfortable with that, and my wife has done the same through her employer.  Since the individual 401k was created in 2004 or 2005, however, I've been allowed to tack on an ever-increasing contribution on top of the 20% -- this year, an extra $22.5k, since I'm over 50.  That is a tall order.  I am always acutely conscious that a large chunk of every allowable dollar that I fail to contribute goes to taxes -- avoidably. So I come as close to maxing out as I can.  I've always assumed that this a good thing -- that this incentive is working as it should.
What dawned on me after reading about this study, which focused on Danish savers since reams of detailed data are available there, was that the incentive doesn't really shape how much I save -- it just controls where I put it. That tracks (roughly) with one key finding of the study in question:

When individuals in the top income tax bracket received a larger tax subsidy for retirement savings, they started saving more in retirement accounts...

... but the same individuals reduced the amount they were saving outside retirement accounts by almost exactly the same amount, leaving total savings essentially unchanged. We estimate each that $1 of government expenditure on the subsidy raised total savings by 1 cent.

About 90% of our savings, excluding home equity, is in retirement accounts. That's not good. Or rather, it's only "good" if you assume that it's in the natural order of things for retirement funds to be especially privileged.

One option we always eschewed was College 529 savings.  The investment options were so crappy and so limited, we just invested savings for college conservatively in our own accounts and paid taxes on quite limited interest income. In fact the 529 plan structure is a scam, the history of which I would like to research one of these days. Have you ever wondered why you're free to invest IRA funds wherever/however you want, while some crappy mutual fund company or handful of companies has monopoly or cartel privileges in the 529 options for your state?  Why, for example, you probably can't put the money in index funds?  There must be a lovely lobbying tale behind the genesis of the 529s. 

What the Danish study tells me is that all savings should be equal, and all citizens should be able to avail themselves of the same limited tax credits to save. And oh yes, we should be free to put those tax-protected savings into whatever investment vehicles we choose.

The other key finding of this study is that "automatic" savings plans are highly effective  -- that is, a system in which employers make it a "default" option, one you have to deliberately opt out of to avoid, to have retirement savings deducted out of every paycheck.  Apparently, when left to their own devices, only high earners will save much at all, anywhere, so boosting the retirement savings of most people remains a worthy goal.  But to fulfill that goal, encouraging automatic savings programs is likely to be more effective than providing outsized personal tax incentives to save.