That statement is false. You are extrapolating wildly from a handful of simplistic laboratory experiments where organisms were removed from their complex natural environments, put in much simpler synthetic environments, where the main mode of selection is unavoidably for replication speed.
The nutrients in the synthetic flask environment is limited, and runs out every day hours before a subset of the population is passed on to a flask with fresh nutrients. Hence the organism’s descendants that can grow and divide the fastest, when it comes time to transfer to a new flask, will have come to constitute a larger fraction of the population than it’s competitors. So the experiment is unavoidably indirectly selecting for reduced genome length.
That means the organism is wasting time expressing and replicating many genes with functions it doesn’t need, hence mutations that inactivate and eventually delete these genes have become beneficial in the synthetic environment(where they were beneficial in the natural one it was taken from).
Every time you repeat your claim there you are stating a falsehood.
Even to the extent that so-called “degenerative” mutations are more likely than constructive mutations, the conclusion doesn’t actually follow. It doesn’t follow that because mutations are more likely to degrade a function/information, than to create or improve it, that evolution can never amount to a net increase in functions/information. You are focused narrowly on a small subset of inherent mutational biases(point mutations and deletions, as opposed to things like duplications) and neglecting to consider the impact that environment has on the process.
First of all there is going to be some floor below which no more degenerative mutations are tolerated. “Degenerative” mutations that occur in needed functional genes will be selected against, hence the process of loss will reach a floor below which no more genetic material can be dispensed with, as all there is left is genes with important functions required to sustain the organism. An organism that loses a gene critical for carrying out the function of replication, won’t be replicating faster than it’s competitors(it won’t be replicating at all), and so will not have come to dominate the population when a small fraction is transferred to a flask containing fresh medium the next day.
In this state, for organisms in the wild where it is not the case that the primary selection pressure is replication speed, since constructive mutations can still occur, though they may be much more rare than degenerative ones, if the function they perform in the environment outweighs the reduction in replication speed conferred from expressing and replicating them, they can become fixed and thus contribute to an increase in genetic information. That is how, for example, you can get novel genes that confer antibiotic resistance. Or already existing genes can be duplicated and low-level promiscuous side-reactions can suddenly become selected for.
While a novel gene might make the organism use more resources on expressing this novel gene, and spend more time replicating this additional gene(supposing it’s a duplication), if this novel gene allows the organism a large growth advantage when facing a novel challenge (such as an antibiotic), it can be selected for and the increase in genetic information can become fixed.
As should be obvious now to a thinking person, the conclusion you are seeking to extrapolate from a handful of simplistic laboratory experiments, doesn’t actually follow. Both reason and evidence shows that genetic information can actually increase under many circumstances even in the face of an intrinsic mutational bias towards “degeneration”. Interestingly, that is how you get constructive neutral evolution, which you can read about here:
1: Stoltzfus A. On the possibility of constructive neutral evolution. J Mol Evol.
1999 Aug;49(2):169-81. PubMed PMID: 10441669. DOI: 10.1007/pl00006540
Stoltzfus A. Constructive neutral evolution: exploring evolutionary theory’s curious disconnect. Biol Direct. 2012 Oct 13;7:35. DOI: 10.1186/1745-6150-7-35
For a concrete example of where scientists have been able to piece together how constructive neutral evolution contributed to increased complexity of a molecular machine(it’s your favorite one, ATP synthase), you can read this:
Finnigan GC, Hanson-Smith V, Stevens TH, Thornton JW. Evolution of increased complexity in a molecular machine. Nature. 2012 Jan 9;481(7381):360-4. DOI: 10.1038/nature10724