Indigenous pig breeds are still dominant in Laos, which covered more than 91 % of the total pig population (3.1 million) in 2014. Most of them were raised by smallholder farmers, in particular, in rural or mountainous areas (Keonouchanh, 2018). Indigenous pigs not only contributed to meat consumption and primary income for households (Holt et al., 2019; Okello, 2017), but also they were also important to the traditional activity performance of farmers, especially ethnic people living in remote areas of Laos (Keonouchanh et al., 2014). In a similar report of Xayalath et al. (2020), indigenous pigs play an important role in the supply of meat and contribute between 9 %–14 % of the annual household income in the rural areas of Laos. It could be said that the ethnic people (Lao-Tai, Mon-Khmer, Hmong-Mien, Tibeto-Berman) living in the mountainous areas of Laos, especially in the north, could not abandon indigenous pig raising (Phengsavanh et al., 2011). In 2019, Laos counted a pig population of about 4.1 million, including 1.4 million in the north, 1.2 million in the central region, and 1.5 million in the south (MAF, 2020). Lao government has been active in encouraging farmers to move from conventional production to farming production, which could both control quality and increase pork production to meet pork consumers' needs (MAF, 2015). However, the southern part had the largest number of pigs (with more commercial pig farms). Still, the northern region also had the second largest pig population with an enormous number of indigenous pigs, almost 85 % of the total pig population. In northern Laos, more than 60 % or some areas (Phongsaly Province), more than 90 % of the total households owned at least one indigenous pig (Epprech et al., 2018). Unfortunately, since 2016, not only in Laos, the number of indigenous pigs has decreased. Because of the African swine fever (ASF) disease and other factors, it has rapidly declined in all Association of Southeast Asian Nations (ASEAN) countries as well. However, indigenous pigs produce delicious and healthy pork, and the new market options should be found for the survival of these breeds and for the poverty alleviation in rural areas. Indigenous pig production in Laos is facing two main chronic problems: sows give a small number of piglets per litter and poor farm management, causing various negative impacts on pig performance. This includes only about 5–8 piglets per litter with a few survival piglets at the time of weaning (mortality rate 20; 50 %) and poor growth rate (Chittavong et al., 2012; Phengsavanh et al., 2010) with an average daily gain less than 120 gd-1, and high fat rate about 65 %–70 % (Keonouchanh and Dengkhounxay, 2017). Such problems, including the quality of productive and reproductive performance, are still a concrete problem that blocks the progression of improving indigenous pig performance in Laos. These issues are waiting to be addressed, especially primarily housing and feeding technology, animal health care, and better marketing options. The fruitful experience of the Mangalica indigenous pig farm operation in Hungary, Black Iberian pigs, and other developed European indigenous pig breeds could be the best example for the world's indigenous pig development model.

We need to find out all the holes in the development of indigenous pigs and find the options of the better practices of Hungarian Mangalica pig farms in Hungary, to be complied with and adopted or recommended to improve the quality of indigenous pig performance for farmers in Laos. This study was conducted with two main objectives. The first was to study the general trend of the indigenous pig population and compare the reproductive performance of indigenous pigs based on clustering villages and rearing systems. We also indicated two hypotheses that indigenous pigs raised by farmers who live near downtown would have better reproductive performance parameters than those raised by farmers who live far away from downtown because of easier access to feedstuffs and other facilities. Another hypothesis was that the indigenous sows raised in the confinement rearing system would have better reproductive performance parameters than the other two raising systems (free-scavenge and semi-scavenge rearing systems), as farmers would provide them with better feedstuffs and other practices.

In the study areas, indigenous pig production was mainly based on traditional rearing systems such as simple confinement, free scavenging, and semi-scavenging and found to be around 39 %, 25 %, and 36 %, respectively. The pig rearing systems depended on the seasons of agricultural production and the village regulations. Most of the villages in the study areas are now striving hard to build and lead their villages to meet the criteria of becoming a sanitized and developed village, with respect to three constructive levels of national development strategy (Central committee party office, 2012). Keeping animals in pens, especially pigs, is also one of the criteria that every village has to consider and follow as a wide-range regulation in many areas. Pig raising as a semi-scavenging rearing system usually comes in two forms. In the first form, farmers kept their pigs in pens only in the night-time after they fed them in the evening. During the daytime and otherwise, they would keep their pigs in pens only during the main crop production season and leave them in free-scavenging mode again after harvest. This is very similar to earlier Hungarian farmers in the villages who kept their Mangalica pigs indoors only in the night-time, and they grazed all their pigs in the communal pasture (Egerszegi et al., 2018).

Pig housing, nutrition condition, and proper selection of both gilts and boars are three main issues to consider for improving indigenous Lao pig breeds' current productive and reproductive performance in northern Laos. Egerszegi et al. (2018) summarized the past and current status of Hungarian Mangalica pig with special regard to breeding and reproductive performance research: (1) select replacement gilts, large litter, and at least 10 functional teats; (2) stimulate gilts and weaned sows to come into heat at a similar time and get piglets in the same batch; (3) improve housing and nutrition in case of intensive breeding; and (4) improve feeding of sows bred intensively and culling after sows have achieved her six–seventh farrowing. These successful development tracts should be introduced to improve the quality of indigenous Lao pigs.

Two different types of local pig breeds – i.e. Moo Lath and Moo Hmong – were found in the study areas. These two types were commonly found in northern Laos. These two breeds were different in the growth rate and reproductive performance, especially litter size of 7–8 versus 7–10 (Keonouchanh et al., 2011). However, no evidence on genetic information was found that described the difference between these two types of pig breeds, but their phenotypes were so different among them (Wilson, 2007). Due to the differences between types of breeds, this may be one way of explaining the difference in litter size and birth weight (Gokuldas et al., 2015). It was found that there was no difference between FC (<15 km) and ThC (>30 km), but the SC (15 to 30 km) had a high litter size and birth weight. However, the SC had a slightly low litter size per year. It may not be surprising that the average weaning weight was higher in the ThC (7.83 kg) than FC (6.86 kg) and SC (6.97 kg). This might because of having a longer period of suckling (2.93 months) compared to FC (2.36 months) and SC (2.43 months). Wilson (2007) also reported that Moolath pigs reached puberty at around 6 months of age, piglets weaned at 2–3 months at around 7–8 kg, and sows had about 1.5 litters per year. Therefore, it was hypothesized that the farmers did not consider using the community feedstuff services and other facilities available in the nearby cities. Most pig farmers in northern Laos mainly used locally available feed by-products from agricultural production and seasonal forest vegetables to feed their pigs (Phengsavanh et al., 2011). This finding slightly differed from the study of Phengsavanh et al. (2010), which concluded that the weight gain of fattening pigs raised by farmers in northern provinces of Laos, who lived near the city centre, was better than that of farmers who lived far from the city centre.

This study's findings may also differ from other theories and research where it was found that many parameters of indigenous pig reproductive performance in the free scavenging were better than the confinement and semi-scavenging rearing systems, especially the litter size, birth weight, and weaning weight. It might be hard to explain with the right reason; again it could be assumed that feeding and scavenging could be one option to consider what made them different. There was no difference in feeding among the three different rearing groups. Farmers always fed their pigs once in the morning and another once in the evening with available local feed resources such as cassava roots, rice bran, and kitchen waste produced from agricultural production (Huynh et al., 2006). Due to the smallholders, farmers could not buy high-quality feedstuffs or completed feed for their pigs (Stur et al., 2002). The pigs in the free-scavenge rearing system were able to consume, including scavenging, and were fed by their owners in the morning and evening.

Almost 51 % of farmers reported that their sows gave birth during the main agricultural production season between June and September. Around 24 % gave birth to the piglets after the harvesting season between October and December and about 25 % during the non-agricultural production season. For most wild boars and earliers Hungarian Mangalica pigs the heat and mating occurred in the autumn and farrowing 4 months later in the spring, and sometimes a second farrowing might have again in August and September (Webster, 2011; Egerszegi et al., 2003, 2018). It is also hard to accurately describe the difference (P<0.038) in the mean of litter per year of the sows, which farrowed during the primary crop production, after harvesting, and non-agricultural production seasons at 1.40, 1.18, and 1.20, respectively. However, there was no difference (P>0.050) in the mean of the litter size, birth weight, weaning weight, and mortality of piglets before weaning for the sows farrowed during main crop production, after harvesting, and non-agricultural production season. But it was found that the piglets farrowed during the main crop production and after harvesting season had a slightly higher birth weight (0.73 versus 0.77 kg) than those born during the non-agricultural production season (0.68 kg). Conversely, piglets were farrowed during main crop production and after harvesting season had a slightly high mean piglet mortality before weaning (1.23 versus 1.14 heads per litter) compared to piglets born during the non-agricultural production season (1.10 heads per litter). It was assumed this was because of improper management during farrowing. Besides, these two seasons, farmers always kept their sows in pens, which could make it easy for sows to crush their piglets.

The mortality rate (15.45 %) was lower than the study of Phengsavanh et al. (2010) with around 50 %, which was carried out in the same region, and the study of Chittavong et al. (2012) with around 20 %, which was conducted in the central provinces of Laos. It could be considered that the piglet mortality might be high for the early weaning due to new transit and improper feeding by farmers. Based on the early weaning period, one of the key aspects to consider for the piglet survival and growth rate is that new transit could increase more stress, and new feed could be harmful to the new transit piglets (Campbell et al., 2013). It was interesting to note that this study found that the FRS had the lowest piglet mortality rate (13 %), compared to CRS (15.91 %) and SRS (16.93 %). The statistics are presented in Fig. 3. It was hypothesized that some farmers might lose the monitoring of their sows for the first three days after farrowing. All households from the FRS group responded that they left their sows to give birth in the nearby forest. Some of them did not even know that their sows had already given birth to the piglets until the sows brought their piglets back home. Again, it was assumed that farmers might not know how many piglets had died before they had been found. Before weaning, most piglet mortality always occurred during the first three days after farrowing, due to weakness, lack of immunity, and the fact that piglets can be easily crushed by their mother (Deviller et al., 2011).

However, the population of indigenous pigs has decreased year by year due to Lao consumers' eating preferences. They increasingly prefer lean (low fat) pork that is provided by exotic pig breeds rather than fatty pork that is characteristic of indigenous Lao pig breeds. Still, indigenous pigs are a key component of meat consumption in Laos. Most farmers raise their pigs mainly under traditional practices, so proximity to downtown or facility services is not crucial for reproductive performance of indigenous Lao pig breeds. In general, the pig rearing systems also do not affect the reproductive performance of indigenous Lao pigs. This study showed that the free-scavenging rearing system had a marginally better litter size and birth weight and had the lowest piglet mortality rate compared with confinement and semi-scavenging rearing systems. Most previous research stated that the free-scavenging rearing system was poor in every parameter of productive and reproductive performance of indigenous pigs compared with other rearing systems. The local community services (e.g. providing feedstuff) and rearing systems have no strong influence on the farmers with special regards to reproductive performance of indigenous Lao pigs. However, proper farm management might have a major impact on overall productivity. Actually, most Lao farmers still raise their indigenous pigs in a traditional extensive system and do not operate their farms in a business manner. Therefore, the combination of the extensive and intensive model of Hungarian Mangalica pig breeding and production system might be a positive option to introduce and adapt to improve indigenous Lao pig performance. Based on the present study, we can conclude that Lao farmers might be able to improve housing and feeding methods and to develop relevant reproductive management of indigenous pigs with the goal of increasing the number of surviving market-weight offspring.