First of all, we have to define what is meant by distorting the sex-ratio.
1) Changes to the sex-ratio as defined by the principles of the underlying Sex-determination system, can be seen as extremely unlikely:
In most species, sex is determined by inheritance of sex chromosomes, which based on meiosis results in a 50:50 sex-ratio. In this system, sex is determined by the presence or exact count of the sex-chromosomes (dependent on the species). Once this sex-system is established, I would personally see it as "frozen" and very unlikely to ever change. Other sex-determination systems are based on temperature and other environmental factors, and you can safely assume that evolution of their sex-ratios is based on advantages in the spread of the responsible genes. (Either through higher transmission rates within each population, or through a higher survival of the populations as a whole.)
2) Distortion of the perceived sex-ratio within the XY sex-determination system via TRD:
Transmission ratio distortion (TRD) occurs when one of the two alleles from either parent is preferentially transmitted to the offspring. This leads to a statistical departure from the Mendelian law of inheritance, which states that each of the two parental alleles is transmitted to offspring with a probability of 0.5. A number of mechanisms are thought to induce TRD such as meiotic drive, embryo lethality, and gametic competition. Importantly, such mechanisms only affect the sex-ratio, if the corresponding selfish gene sits on a sex-chromosome.
In this context, it is important to distinguish between true meiotic drives and killer meiotic drivers, as can be read in this review-paper. Killer meiotic drives often use complex [poison|antidote] systems, in which a tightly joined pair of a killer and an antidote gene together cause lethality in (haploid) gametes lacking the antidote containing chromosome after meiosis, while the poison is still present.
In general, drive alleles are predicted to be transient and evolutionarily labile. The transmission advantage enjoyed by drivers can allow them to become fixed in a population. After fixation, all individuals will be homozygous for the driver and exhibit no drive phenotype. However, if it affects a sex-chromosome, it could cause a bizarre state: On the x-chromosome, the mechanism would create a disadvantage in propagation due to the lack of male off-springs. On the y-chromosome, it might theoretically eradicate the whole population. By the way, such systems are considered for mosquito control.
Importantly, I would also like to address the gene R2D2 is an example of the true meiotic (see original publication), which is a “selfish” genetic element that exploits asymmetric female meiotic cell division to promote its preferential inclusion in ova (while the other female haploid cells die). However, against the suggestion of a commenter, this gene is NOT able to cause a distortion of sex ratio in mice, as females are homologous in X chromosomes; the ova always contains an x-chromosome.
So I will answer your question regarding selfish-genes, if it "isn't always in their interest to cause a biased sex ratio" with a clear NO! Distortion of the sex-ratio is an extremely rare event that can result in strong disadvantages for the responsible gene, causing its own distinction.