Fewer DNA Matches in the Future

For genetic genealogists, the promise of new DNA matches is both exciting and a necessary component of our research strategy. However, a recent demographic study finds that the number of living kin for the average person will reduce by 38% in the future suggesting we may have fewer DNA matches.

The demographic shifts are not immediate, but in some countries and regions, they are already visible. The kinship decreases do not render DNA testing ineffective now or in the future. Rather, the study impresses upon us that we need to become more proactive and thoughtful with DNA testing. In this blog post, I discuss the study’s major findings and its potential implications for genetic genealogy.

Global Kinship Decline
The demographic research study, published in December 2023 in the Proceedings of the National Academy of Sciences (PNAS),[1] finds lower and delayed fertility will drive the projected decline in the number of our future kin. The decision to have fewer children (lower fertility) and children later in life (delayed fertility) creates smaller family sizes in the present which become more pronounced in the future. Additionally, longer lifespans coupled with delayed fertility creates larger age gaps between generations (longer generations) and increases the likelihood that generations span from great grandparents to great grandchildren.

The image below, which is taken from the 2023 study, highlights for a few select countries the number of living kin at a given year (on the horizontal axis) for a woman aged 65. Kin are defined as great-grandchildren, grandchildren, children, nieces/nephews, siblings, cousins, parents, aunts/uncles, grandparents, and great grandparents. Of all countries, Zimbabwe and Italy represent the countries with the largest and smallest family sizes in 1950, respectively.

While smaller family sizes are already found in Europe and North America, the most dramatic decreases in family sizes by the year 2095 will be in Latin America and the Caribbean. Yet, all regions will experience further declining family sizes in the future. Interestingly, we only need to look at the effects of China’s one child policy (1979-2015) to envision what our future may look like. For a Chinese woman born in 1950, 39% of her kin were categorized as cousins, but fast-forward to 2095 and only 7% of her kin will be cousins.[2]

Although the implications derived from the PNAS article were principally related to social support, the research is generalizable to the world of genetic genealogy. The above graphic assumes a woman aged 65, but evidence suggests that the average age of someone interested in genealogy, and thus likely to initiate DNA testing for genetic genealogy, is in their 40s and 50s and getting younger.[3] So, rather than look at data for a woman aged 65, it is perhaps more appropriate to view the study’s other data for a female aged 35-39 years old.[4] To provide greater contextualization, the study’s two extreme year end points of 1950 and 2095 for a U.S. woman is used, and her living kinship is visualized as a family tree, which is perhaps more relatable to genealogists.

As visualized above, the number of living kin from 1950 to 2095 becomes less horizontal and more vertical. It resembles less of a Christmas tree in form and more like an Italian cypress. In fact, the 2095 family tree contains 50% fewer individuals, and generational gaps are longer.

How Living Kin Affects Genetic Genealogy
There are four current realities about genetic genealogy suggesting that smaller family sizes will result in fewer DNA matches in the future. With respect to commercially available autosomal DNA tests like Ancestry, FamilyTreeDNA, and MyHeritage, the first reality is that we can only test living people. Our ancestral DNA is randomly inherited, and so not all family members have the same DNA matches. This is why genetic genealogists test more than one family member hoping to find at least one match in a cousin’s list of matches that breaks down our brick wall. With smaller family sizes, we have fewer living cousins to test and fewer chances for anyone of them to share DNA with our mystery ancestor.

The second reality is that currently autosomal DNA testing can only reliably predict genealogical relationships up to the second cousin level.[5] For adoptees or recent non-paternity events (NPEs), smaller family sizes may not be any more challenging than it is today. In fact, provided close matches are found, smaller family sizes may make it easier for adoptees to identify parents because families have fewer children to hypothesize parentage. Yet, finding close matches will be more challenging because smaller family sizes mean there will potentially be fewer close matches to begin with.

However, for those of us searching for ancestors at the fourth cousin or greater level, it is recommended that 10 or more first and second cousins match the potential distant cousin to have enough data to statistically conclude the genetic relationship.[6] With smaller family sizes, we will have fewer first and second cousins to help us prove the relationship. The genetic networks of shared matches that have proven so helpful to us today will become smaller in the future. It therefore becomes more difficult to reliably reach further back in time to identify our more distant ancestors.

For paternal (Y-DNA) and maternal (mitochondrial DNA) testing, the third reality is that we need either an unbroken male to male or female to female generational relationship to determine if individuals share parental or maternal ancestry, respectively.[7] Often, these tests are used to confirm relationships several generations back in time. With smaller family sizes consisting of one to two children per generation who may or may not have had children of their own, the likelihood that genealogists find an unbroken male or female line will become exceedingly more difficult than it is today.

The fourth reality is that most of our DNA matches (especially autosomal DNA) depend on the random decisions of our cousins to take a test. Because most of our matches are distant cousins (5^th to 8^th cousin, or 6-20 cM in length), this pool of matches will decrease over future generations as our smaller group of 2^nd and 3^rd cousins move to the 5^th and greater cousin level as younger generations test. While our closer matches (below 5^th cousin) have typically constituted the smaller share of our matches, these will shrink, too, as family sizes become smaller.

How to Genealogists Can Mitigate the Effects of Smaller Family Sizes
While the people comprising our cousins are not random, those taking DNA tests are mostly random. Anecdotally, most of the DNA matches that have put me on the research path toward breaking down brick walls were cousins randomly appearing in my match list. Yet, it was the subsequent collaboration and targeted DNA testing that ultimately confirmed the break-through. To mitigate the effects of smaller kinship in the future, genetic genealogists need to be more proactive and forward-thinking in our approach in what I call the three Cs (3Cs) forming the future of genetic genealogy: coverage, collaboration, and contingency planning.

Coverage. Especially for autosomal DNA testing, two truths are evident: DNA inheritance is random and older generations have larger segments of our ancestors’ original DNA. For example, one sibling could inherit a 24 cM segment from their 2x great grandfather while another sibling does not. If we only tested the second sibling, we may not discover the group of shared matches (i.e., genetic network) leading us to the identification of our 2x great grandfather. In the future with fewer siblings, aunts/uncles, and cousins, the random luck in finding the 24 cM segment, or a large enough segment to be genealogically relevant (i.e., greater than 15 cM),[8] will become more challenging because, with fewer births in the family, the segment may not be passed down at all.

So, it’s important for genetic genealogists to test as many of their living kin today as possible with a greater emphasis on older generations. Doing so not only helps with research now, but in the future as discussed with contingency planning shortly. Many of us know this, but are we actively doing it?[9]

Just to put a finer point on coverage, consider the following example. I have access to 22 kits on Ancestry.com who are descendants of my 4x great grandparents John Wilson (1784-1840) and Elizabeth Boyd (1784-1860). Before I knew Elizabeth’s maiden name, I found numerous matches forming several genetic networks that ultimately led me to identify her parents. The table below lists each of my cousins’ kits anonymized numerically from 1 to 22 and grouped under each of John and Elizabeth’s five children from which my cousins descend. Six select Boyd matches are exemplified in the table and anonymized alphabetically from A to F. Table entries list the size of the of the match in centimorgans (cM). Note how matches are not equally distributed across all descendants thereby stressing the point that DNA inheritance is random and that by maximizing coverage, the identification of relevant DNA matches becomes more likely.

Collaboration. To exploit and extrapolate on the familiar saying, “it takes a village to find an unknown ancestor”. Of the 22 kits used in the above example, 68% represent cousins who shared their list of DNA matches with me (I manage the other 32% of kits). The genealogical community is good with sharing information and responding to others’ queries, but in the future, we’re going to have to do better because many who take DNA tests are simply curious about their ethnicity and not generally interested in genetic genealogy. These matches often don’t respond to our messages, and as a community, we need to figure out how to engage these matches and make them care about genetic genealogy.

Furthermore, with smaller family sizes, family data and stories are not going to be as widely distributed as they perhaps once were. It will be less common to hear, “Oh, I think I have a cousin who might know something about that.” We need to not only document family histories but be more willing to share them.

I’ve personally experienced individuals who did not want to share information because they worked so hard to uncover it. They want to protect their investment in time and perhaps money. Yet, we know that the answer to one genealogical question creates several more questions needing answering. Sharing discovered information in a very public way enables others to pick up where you left off. This is why I publish all my research reports on my website. I believe there is an opportunity for companies like FamilySearch and Ancestry.com, who have the necessary resources, to champion collaboration in ways that we (and they) yet can’t envision that are even more egalitarian and not behind paywalls. These companies have made great progress here, but we need to do more especially with outreach and motivating genealogists to participate.

Contingency Planning. We don’t like to think of our own mortality, but this is exactly what we must do if we are to ensure our efforts to document our family history is not buried in a file on computer destined to be reformatted in preparation for its next owner. If you’re like me, I tell myself that I will tackle the inheritance of my research when I solve just one more family mystery. Only then will I have time to think it through. Well, that proclamation was made many years ago and certainly a dozen solved mysteries ago, too.

Sustained declines in the number of our kin and thus fewer DNA matches in the future emphasizes the importance of the DNA matches we have now. Indeed, our current DNA matches likely have more shared matches than the ones we will have in the future. We need to preserve these present-day connections so that future genetic genealogists can tap into these genetic networks to solve family mysteries that we have yet to identify or could not yet solve.

With shrinking family sizes, it will be increasingly difficult to find a niece, nephew, or cousin interested in the computer files and boxes of paper we’ve accumulated over time let alone sharing access to our online genealogy accounts where countless other saved work resides.

I believe the industry is in desperate need of an entity willing to champion or archive our work. Yes, some historical societies are willing to take portions of our research if it is relevant to their constituents and in a coherent and organized form, but even their resources are limited. At the very least, genetic genealogists need to consider doing the following:

1. Identify a Beneficiary. For your DNA and/or your kin’s DNA that you manage, designate a beneficiary who will inherit the ability to analyze and curate genetic discoveries in the future. Your beneficiary can be stated in your will or trust. Alternatively, some DNA testing companies let you designate it in your account. For example, FamilyTreeDNA has a beneficiary tab in your account settings where this information can be communicated. Ancestry is a bit more complicated in this regard but does specify in an article how beneficiary designation can be accomplished. Your beneficiary could be another family member who has an interest in genetic genealogy or another researcher for whom you’ve built a good relationship with over the years. In either case, a conversation making them aware of your wishes and their willingness to consent is warranted.
2. Consider Making DNA Information Publicly Available. Depending on your comfort level and your perspective of the future, genetic genealogists should link their DNA results to themselves in their family tree, make their family tree publicly viewable (or at least publicly searchable), and upload or test their DNA on multiple DNA sites. Such efforts will not only increase collaboration, but it will ensure your data is available to others without the necessity to contact you (especially if you are deceased). Most of us do this, but our community and the companies operating in this space can and should do more to encourage others to make their DNA results more public including investing more in data security and engaging in outreach efforts.

Summary
This thought piece on the future of genetic genealogy and DNA matches resulting from the sustained declines in family sizes is not meant to startle or insinuate that the sky is falling. Rather, the intent is to inspire and motivate genetic genealogists to do more now and to ensure others (and themselves) can do even more in the future. To ensure the large legacy of our work, I encourage genealogists, and the companies that support us, to discuss and strategize how a sustained decline in family sizes will influence genetic genealogy. Comments are welcomed.

---

---

Sources
[1] Alburez-Gutierrez, Diego, Ivan Williams, and Hal Caswell (2023). “Projections of Human Kinship for All Countries,” PNAS, 120 (52), 1-6, retrieved from https://www.pnas.org/doi/10.1073/pnas.2315722120.
[2] Ibid.
[3] Shaw, Emma (2019). “Who We Are, and Why We Do It: A Demographic Overview and the Cited Motivations of Australia’s Family Historians,” Journal of Family History, 45(1), 109-124; And, Zippia Inc. (2023, July 21). Genealogist Demographics and Statistics in the US, retrieved January 15, 2024 at https://www.zippia.com/genealogist-jobs/demographics/.
[4] Data taken from p. 61 in the appendix of Alburez-Gutierrez et al.’s (2023) study, retrieved January 14, 2024 at https://www.pnas.org/doi/10.1073/pnas.2315722120#supplementary-materials. To create “whole” people, fractions were rounded up or down to the nearest whole number. To simplify the display, aunts/uncles and cousins are only shown for one parent’s perspective. Displayed years take the first year in a range, e.g., 1950 for 1950-1955.
[5] International Society of Genetic Genealogy (2020). Autosomal DNA, retrieved January 14, 2024 at https://isogg.org/wiki/Autosomal_DNA#.
[6] Ibid.
[7] Note that while men inherit mitochondrial DNA from their mother, they do not pass it down to their offspring.
[8] Bettinger, Blaine (2017), The Danger of Distant Matches. Accessed 10 October 2023 at https://thegeneticgenealogist.com/2017/01/06/the-danger-of-distant-matches/.
[9] Note that if you can test both parents, it’s not necessary to test their children as their children’s DNA is fully represented by both parents.

---