The most extensive and authoritative report … concluded that 2 percent of men with “high paternity confidence” — married men who had every reason to believe they were their children’s father — were, in fact, not biological parents. Several studies indicate that the rate appears to be far higher among unmarried fathers. …

At a federally convened symposium on the increase in paternity questions, a roomful of child-welfare researchers, legal experts, academics and government administrators agreed that much pain could be avoided if paternity was accurately established in a baby’s first days. Several suggested that DNA paternity tests should be routine at birth, or at least before every paternity acknowledgment is signed and every default order entered.  …

The same care that hospitals take ensuring that the right mother is connected to the right newborn — footprints, matching ID bands, guarded nurseries, surveillance cameras — should be taken to verify that the right man is deemed father.

More here, and HT to Roissy, who supports mandatory paternity testing at birth, as do I. After all:

Most states … have their own mandatory newborn screening programs … Almost all states now screen for more than 30 disorders.

Most of those disorders are much rarer than 2%, and we have a far stronger reason to expect market failure for paternity testing than for the other required tests.  Men are clearly reluctant to request a paternity test at birth because doing so sends a bad signal:

Most of the [paternity acknowledgment] forms do note that genetic testing is available.  Advocates on both sides of the issue, however, say nearly all men sign the form without undergoing testing.  Sometimes they believe they are the father; sometimes they don’t understand what they’re signing; sometimes they hesitate to question a girlfriend’s fidelity right after she’s given birth; and sometimes they sign knowing full well the child isn’t theirs.

So what are best arguments on the other side?

It is, Smith says … a double standard that allows mothers and caseworkers to use DNA to prove paternity but prohibits men from using that same evidence to escape its obligations. But child-welfare experts counter that a child shouldn’t be punished by losing the only father she has ever known — or the financial security he offers — just because he’s upset that she doesn’t share his genes. … Child-welfare advocates say that making biology the sole determinant of paternity in cases like Smith’s puts the nonbiological father’s interest above the child’s. … ‘Is it my kid?’ is irrationally important to the cuckolded husband,” says Carol McCarthy … of the American Academy of Matrimonial Lawyers. … Why is it that we imbue genetic relationships with a potency that borders on magic?

My wife added that mandatory baby paternity testing seems to unfairly punish moms who “accidentally” misled men about paternity, and it would legitimize harmful suspicious attitudes of men during pregnancy.

I find these contrary arguments to be extremely weak.  If we worry about the kid’s financial security, why don’t we tax everyone instead of just this one man?  What basis could we have for calling this man’s reluctance “irrational,” if the rest of us are equally unwilling to pay for this kid?

Biologically, cuckoldry is a bigger reproductive harm than rape, so we should expect a similar intensity of inherited emotions about it.  If 2+% of women were raped and we had a reliable cheap way to identify the guilty party, don’t you think we’d require that?

Added 14Dec: More rate data:

More than 10% of births in the U.S. are to women with more than one concurrent heterosexual partner. … 20% of births are to women who have had more than one sexual partner in the previous year. … Over-all men falsely believe to be their own biological children about 4% of children.