Here's the case against them.
First, it is fairly easy for large producers of homogeneous products to add nutritional labels to their products. The one-off testing and label re-jigging is a fixed cost that is spread across a very large number of units. But, suppose you're a craft brewer and you get a notion to make a seasonal autumnal ale with pumpkin in it. It'll taste good and sell well as a small batch. But after you make a batch, you're going to have to send a bottle to the lab for testing, wait for the results, and attach the appropriate nutritional label to your new brew. This will add maybe a fortnight or more to your brewing cycle on the first batch of the product and cost you a bit in testing. You can't spread those costs over many units because you're not making many units. And heaven help you if you decide you should double the pumpkin in the next batch.
Second, it's a non-tariff barrier against imported products made in countries that do not have labelling requirements. When New Zealand implemented labelling requirements for standard drinks a few years ago, the shop where I bought my oddball foreign grey market craft beers had to print off little labels for each bottle of the one-offs that they sold, converting the percent alcohol content into a number of standard drinks. This added to the cost of foreign craft beer relative to domestic or mass market product.
You will rightly note that this is also an argument against mandatory nutritional labelling requirements on any small volume products.
If there's any steam behind nutritional labelling requirements, there are things we can do to make it less awful.
The easiest labelling requirement would only require that producers give a general range of calories contained per serving of the product based on the alcohol content alone. A gram of alcohol has seven calories, so a standard drink contains 70 calories. Most products could then simply say something like:
"One serving of this product provides 50-100 calories through its alcohol content."