The Myth of the Investor Class

Although the ownership of stocks and bonds is more highly concentrated than ever, we've been hearing a great deal lately about the rise of an "investor class." This concept, used with much abandon by free-market theorists and political operatives, holds that the simple act of participating in the stock market, even if indirectly and in only small amounts, moves people into this new investor class -- a class that supports as little regulation and taxation as possible, because that benefits America's companies and therefore the stock market.

How plausible is this? Should free marketeers really be cheered? Should advocates of social outlay, regulation and progressive taxation be alarmed?

We can distinguish between two versions of the theory. In the shorter-acting version, as more voters own stock, they turn toward Republicans and Democratic conservatives to defend and extend their interests as stockholders. In the longer-acting version, as more voters own stock, they become increasingly skeptical of government regulation, the welfare state and social insurance. Instead, they become increasingly enamored of minimal taxation and individualist approaches to issues such as funding retirement and securing health insurance. Politics in general moves right to accommodate these views.

One aspect of both versions is true. Increasing numbers of Americans do hold stock. About half of today's households qualify as stockholding in some form. That's up from 32 percent in 1989, a considerable change in a relatively short period of time.

And voters specifically are more likely to own stock than households in general. According to the 2000 Voter News Service (VNS) exit poll, 70 percent of voters owned some stock, reflecting the fact that voters tend to be from more affluent households.

But a closer look reveals that about one-quarter of stockholding households own stock worth less than $5,000, and that the typical stock-owning household in the middle of the income distribution (between the 40th income percentile and the 60th income percentile) owns only about $15,000 in stock. Most stock-owning households -- about 60 percent -- hold stock only indirectly, generally in tax-favored retirement accounts (where, of course, they already pay no tax on dividends received). In short, the majority of households lumped in with the investor class only have small stock holdings, mostly in retirement accounts.

It doesn't make much sense to speak of the investor class as a discrete group when it's so incredibly broad, encompassing up to 70 percent of the electorate. Groups only get this broad when they include, say, Bill Gates and a family with a $50,000 annual income and a $15,000 401(k) account. You can do this, but does it tell you anything really useful about politics? In particular, does it tell you anything you couldn't learn from knowing a voter's income level? Does a voter with a $50,000 annual income and a 401(k) worth $15,000 really differ politically from a voter with the same income level who lacks that 401(k)?

With these questions in mind, let's first test the shorter-acting version of the theory: Are these so-called investor-class voters already voting differently than their noninvestor counterparts? According to the 2000 VNS exit poll, those who held at least some stock voted for George W. Bush by 51 percent to 46 percent; those who did not hold stock voted for Al Gore by 52 percent to 44 percent. That's hardly overwhelming, though at least in the "right" direction for the theory. However, it turns out that a good deal of this relationship is accounted for by the fact that voters with higher incomes, who are more likely to vote Republican to begin with, are also those more likely to own stock. Once you take that relationship into account, most of the investor-class effect disappears.

Take our voter with the $50,000 annual income and the 401(k) worth $15,000. The VNS suggests that such voters had a slight preference -- 51 percent to 47 percent -- for Bush in the 2000 election. But the VNS also indicates that voters with the same $50,000 income but no investments preferred Bush by the same margin. Because these voters are in the income group where stock ownership has gone from fairly uncommon to the norm, this is a particularly surprising result. Apparently moving into the investor class, by itself, doesn't make much of an impact on these voters, at least initially.

Perhaps it's white middle-class voters among whom investor-class trends are having their real effect. This is, after all, the real source of Republican voting strength in the last several elections. But the VNS tells us that while white voters with $30,000 to $75,000 in annual income and some investments favored Bush by 56 percent to 41 percent, white voters at the same income level lacking investments preferred Bush by a nearly identical 56 percent to 42 percent. The secret of recent Republican dominance among white middle-class voters clearly lies elsewhere.

Looking at the data by age reveals another interesting pattern that throws the shorter-acting version of the investor-class thesis into question. Among whites 18 to 64 years of age (72 percent of all investor voters), being a member of the investor class had absolutely no effect on one's likelihood of voting Democratic. But among whites age 65 and older, noninvestors were 9 percentage points more likely to vote Democratic than their investor counterparts. A reasonable hypothesis is that seniors who lack investments tend to be less economically secure and more dependent on Social Security income, and therefore more likely to vote Democratic. In other words, it's less that investor seniors love the market and American businesses than that noninvestor seniors are exceptionally economically insecure. But that dynamic doesn't have the same force among voters not drawing retirement income, so we don't see much difference between investor and noninvestor voters under age 65.

This initial foray into the data suggests a couple of things. First, it does not appear that the spread of stockholding among voters is paying immediate and large dividends to the Republican Party. So the shorter-acting version of the investor-class theory does not appear to be true.

Second, the inevitable and strong relationship between stockholding and income level suggests a need to modify the investor-class terminology. As a single, overarching category, independent of income level, it obviously obscures more than it clarifies. Let's substitute instead the idea that there is a large group of middle-income voters -- call them the "401(k) class" -- who own moderate amounts of stock, typically in mutual funds in retirement accounts, and are dependent on these accounts to fund their retirements. This is a large group also (perhaps 30 percent of households, though data limitations preclude an exact estimate), but it permits us to cut out the poor, who tend to have very small holdings (remember, one-quarter of stockholding families have holdings of less than $5,000), the genuinely affluent, who tend to have fairly large and diverse stock holdings, and the flat-out rich, whose status in life is very far indeed from those in our 401(k) class.

This slimmed-down concept, besides having a more sensible size, has a singular virtue: It eliminates the conceit that someone holding $300 million in stock in 61 companies (as Treasury Secretary John Snow did before he was confirmed) and some FedEx worker holding $30,000 in stock funds in her 401(k) are somehow members of the same "class." Instead, it allows us to focus on a group that might actually have some coherence of economic interests and for whom stock investments are means to very specific ends, such as a secure retirement.

Now let's use the 401(k) class concept to test the longer-acting version of the theory. Maybe the 401(k) class isn't voting Republican today. But perhaps its stake in the stock market is leading it toward the kind of anti-tax, anti-spending, anti-regulation agenda that will eventually translate into increased support for free-market candidates and free-market policies. Data from the 2000 VNS exit poll allow us to test this proposition.

First, the VNS asked voters whether government should do more to solve problems or if government is already doing too many things better left to businesses and individuals. Just 42 percent of the 401(k) class (defined here as between $30,000 and $75,000 in annual household income and holding some stock, either directly or indirectly) said the government should do more. But then that was 5 percentage points more than the 37 percent of middle-class noninvestors who wanted more government problem solving. Clearly there's not much of an investor effect there.

Voters were also asked whether they wanted the next president to continue Bill Clinton's policies, change to more liberal policies or change to more conservative policies. Fifty-three percent of the 401(k) class said they either wanted to continue Clinton's policies or change to more liberal policies. Fifty-four percent of middle-class noninvestors agreed.

In terms of tax cuts, voters were asked whether they preferred a larger tax-cut plan, providing an across-the-board tax cut to most people, or a smaller tax-cut plan, targeted mainly to lower- and middle-income people. Forty-seven percent of the 401(k) class supported targeted tax cuts, just under the 49 percent of middle-class non-stockholders who did so.

Another question looked at voter sentiment on the best way to provide senior citizens with prescription drugs: increase funds for Medicare or provide funds for private insurance plans. Among the 401(k) class, 59 percent preferred funding through Medicare; among middle-class noninvestors, 61 percent opted for the Medicare approach.

The survey also asked whether, when a public school is failing, the federal government's priority should be to fix a school's problems or help parents to pay for private school. An overwhelming 85 percent of middle-class non-stockholders endorse fixing failing schools' problems -- and so do 82 percent of the 401(k) class.

The 401(k) class and middle-class noninvestors do show some differences on the issue of whether individuals should be able to invest some of their Social Security taxes in the stock market. This general question, where no hint of possible risk or benefit cuts is mentioned, predictably elicits a high level of support from the 401(k) class (63 percent), somewhat higher than the 56 percent support among the rest of the middle class and just a bit higher than among all voters (61 percent).

However, Social Security privatization inevitably requires a cut in guaranteed benefits. When this is pointed out, support among all voters for Social Security privatization is only in the 20 percent range. This implies that solid support for Social Security privatization among the 401(k) class is probably only a few percentage points higher -- perhaps 25 percent at the most. In other words, not much to build an anti-government movement on.

Furthermore, when voters were asked their top priority for the federal budget surplus (ah, those were the days!), just 32 percent of the 401(k) class chose cutting federal income taxes, compared with 40 percent who chose strengthening Social Security. And, in the 2000 National Election Survey, when voters were asked whether they approved of a proposal to use most of the federal budget surplus to shore up Social Security and Medicare, 80 percent of the 401(k) class approved.

All this does not suggest that the 401(k) class, by virtue of its stake in the stock market, is adopting an ideology supportive of school vouchers, privatized social insurance and other individualist approaches. That is still a political fantasy of free marketeers, not a reality on the ground.

Instead, the 401(k) class is much like the rest of the middle class: concerned about the schools, the health-care system and other problems and perfectly willing to support substantial government action if convinced it is necessary to solve these problems. And this willingness should logically extend to the very area -- retirement funding -- that is primarily responsible for the rise of the 401(k) class. A 401(k), after all, is a creation of public policy. The 401(k) class, not to mention the rest of the middle class (and the poor), would very much like ways to save more, and more efficiently and securely. As noted, the typical member of the 401(k) class has modest holdings and could use a much more solid financial base for retirement.

Indeed, given the recent, sordid history of Enron, the schemes of many other large corporations to underfund their pensions and the failure of accountants to protect shareholders, the theory of the investor class can almost be inverted. As all but the richest investors become more politically conscious of their interests, it becomes clear that they depend on competent government regulation and oversight so that their pension savings are not fleeced, as well as social insurance to protect them from the vagaries of the marketplace in which they invest. The 401(k) class could be a new constituency for the mixed economy.

One thing government can do is mandate a universal pension system that would provide every worker with a fully portable retirement account. Under such a system, a version of which has been introduced by presidential contender and former House Minority Leader Dick Gephardt (D-Mo.), workers could direct cash from any IRA or 401(k) accounts into a comprehensive, fully portable retirement account. In addition, every American child would receive $500 to open an account.

Former White House economic adviser Gene Sperling has advocated a plan to fund these universal accounts further by providing up to $1,000 a year in matching contributions for savings deducted from paychecks -- a 1-to-1 match for middle-income workers and a 2-to-1 match for lower-income workers. The funding for these matches could be generated simply by increasing the threshold for the estate tax (to $5 million) rather than eliminating it entirely and in perpetuity, as the Republicans wish to do.

These reforms would build an investor class, but in a fashion with a profoundly different implication for politics, ideology and policy. Instead of promoting stock ownership in a way that reinforces an individualist approach to government, as Social Security privatization would do, these reforms would give the government a key role in guaranteeing that the promise of stock ownership is democratically extended to all and actually fulfills its purpose of providing economic security.

John Snow and the other very wealthy people lumped in the imaginary investor class might not be too excited about these proposals. But you can bet the 401(k) class, as well as lots of other moderate- and low-income voters, would be. Politicians looking to cement the alliance between small investors and affirmative government, please take note.