A frightening op-ed in The Forward explains that the intermarriage stats, bad as they already sound, might actually disguise the extent of the intermarriage problem:

Imagine there are only four Jews in America, Bob and Carol and Ted and Alice. Let’s say that Bob and Carol intermarry and Ted and Alice in-marry. Since two of the four Jews intermarried, the “individual” intermarriage rate is 50%. But how many couples were created? Bob and Carol both married non-Jews, creating two households. But Ted and Alice married each other, because an in-marriage requires two Jews,
which creates just one household. The result is three households total, with two intermarried and one in-married, or an intermarried couples proportion of two-thirds.

When intermarriage is explained as “almost half the Jews are intermarrying” — in other words, just offering the individual rate for what’s been happening in the United States for the past quarter-century — the word “half,” as huge as that may seem, actually serves to mask the results. The reality on the ground is that nearly double the number of intermarried households has been created compared to in-married households.

Even if the official 'intermarriage rate' is below 50 per cent, then, you might have a situation where the majority of children being born to Jewish parents are being born into intermarried households.